►
From YouTube: AWS UG OSTech Conf 2021 - Day 1
Description
The AWS UG OSTech Conference 2021 is a sprint immense event as the name suggests, a wide event that will have all the developers (SDEs), Cloud, DevOps, SRE, students, and industry enthusiasts all at one place with Interactive sessions by keynote speakers and a real-time application for everything.
Event website: https://www.awsugjaipur.tech/
Agenda: https://www.awsugjaipur.tech/#schedule
We will be having awesome swag opportunities for the attendees who are active throughout the event and also having some contests and quizzes during the event.
1. Quizzes: After each session participate in the quiz and grab a chance to earn a 250/- Amazon gift card.
2. Knowledge Share Contest: Check details on the website.
Submit entry here: https://lnkd.in/gXUjKgM
--
MapMyIndia Contest Submission: https://lnkd.in/g95hB32
--
Surprise Contest from Agora (SDK Integration)|
Top winners will get exiting swags:
Ref: https://docs.agora.io/en/video/start_call_web
Submit here: https://bit.ly/3tO2xjI
--
𝐖𝐚𝐧𝐭 𝐭𝐨 𝐛𝐞 𝐚 𝐦𝐞𝐦𝐛𝐞𝐫?
𝐅𝐨𝐥𝐥𝐨𝐰 & 𝐂𝐥𝐢𝐜𝐤 𝐨𝐧 𝐑𝐞𝐠𝐢𝐬𝐭𝐞𝐫: https://lnkd.in/dNcQKZU
𝐎𝐭𝐡𝐞𝐫 𝐒𝐨𝐜𝐢𝐚𝐥𝐬: https://www.awsugjaipur.tech/joinus
A
A
A
A
A
A
A
B
We
are
so
happy
to
have
you
here,
as
your
name
suggests,
who
is
techconf,
so
it's
not
mean
like
we
are
only
focusing
on
AWS
or
AWS
Services
kind
of
things.
Okay,
so
now
we
are
coming
with
a
broader
level
of
technology,
so
the
name
suggests
Uber
stack
means
open
source
technology,
so
it
can
be
anything
like
devops,
Cloud
native
software
development.
It
can
be
anything
where
you
have
interest
in
okay,
so
you
can.
You
will
be
finding
sessions
in
our
this
conference
in
upcoming
two
days.
B
Okay,
so
there
will
be
lots
of
session
today
also
and
tomorrow.
Also,
if
you
need
any
of
the
information
just
check
on
her
website,
it's
all
here,
if
you
have
any
doubts
about
quizzes
like
contests
okay
or
not
a
share
contest
so
just
check
on
our
website.
It's
all
there.
Okay,
so
I
think
we
are
good
to
go
and
I
have
one
thought
to
just
motivate
you
all
like.
So
the
best
preparation
for
tomorrow
is
doing
your
best
today.
D
B
Yeah,
so
thank
you
all
for
your
honoring
response.
Okay,
so
for
this
conference
we
got
20K
registrations,
okay
and
in
just
three
weeks.
Okay,
so-
and
you
can
check
this-
is
that
you
all
are
there,
so
you
must
be
knowing
okay,
so
the
reach
of
it,
and
now
we
are
having
the
family
of
2500
of
people
in
our
community.
Okay,
it's
it's
growing,
so
it's
kind
of
2500
plus,
but
yeah
yeah.
Next.
B
C
Okay,
so
today
we
have
a
a
great
line
of
speakers,
so
the
keynote
will
be
started
by
Ricardo
and
following
up
with
that,
we
have
pratik
narang
from
Google
and
Michael
Frederick
from
gitlab
Bill.
That
will
be
happening
tomorrow
and
there
are
so
so
many
great
speakers
in
this
conference
yeah
next
slide.
C
B
Yeah,
so
let
comes
to
the
part
where
you
all
must
be
excited
and
again
check
the
comment
section.
So
I
got
a
lots
of
comments
over
this,
so
yeah
we
have
like
in
the
past
some
days.
We
are
having
this
contest.
That
was
live
okay,
but
then
we
have
just
closed
this
because,
as
we
mentioned
in
guidelines
like
we'll
be
having
200,
the
first
user
will
be
getting
the
first.
One
user
will
be
getting
some
goodies
of
specs
from
us,
but
that
deadline
exceed
and
that's
why
we
actually
lose
that
particular
thing.
B
Okay,
but
still
if
you
have
submitted
the
link
or
the
form,
then
you
will
be
reaching
out
to
you
after
25
okay.
So
there
is
no
need
of
asking
this
again
and
again
what
is
about
like
we
have
done
this
particular
contest
or
something
so
this
answers
your
question,
I
think
and
yeah
so
yeah.
So
now
that
is
thing
done,
but
now
also
we
are
having
two
live
contests
in
this
particular
conference.
I
can
say
so
there
will
be
live
with
this
okay.
B
So
after
each
session
we
will
be
having
live
quizzes
for
you.
Okay,
so
you
don't
have
to
go
anywhere,
just
don't
track
on
slack
or
something
like
where
you'll
get
the
links.
You
will
find
the
link
in
comment
section
itself.
Okay,
so
let's
say
if
one
section
ends,
so
you
have
to
join
that
link
and
that
will
be
one
quiz
we
have
planned
for
you
and
you
have
to
just
participate
in
that.
Just
keep
in
mind
check
the
guidelines
of
website.
B
If
you
have
any
queries,
okay
for
the
quizzes,
okay
and
there's
a
knowledge
share
context.
So
what
do
you
mean
by
all
share?
So
let's
say
you
are
attending
the
particular
conference
of
session
and
you
are
having
like
you
are
learning
something
then
what
you
have
to
do.
You
have
to
just
take
a
screenshot
just
post
it
on
your
social
media,
LinkedIn
or
Twitter.
Okay,
just
post
it
there.
If
you
and
we
have
given
you
the
form,
link.
B
Okay,
so
in
the
form
link
we
have
already
given
there
one
link.
Okay,
you
can
submit
multiple
links
there.
If
you
are
posting
at
the
multiple
platform
and
tag
us
and
tag
the
speaker
from
which
you
are
learning
and
what
you
have
learned,
so
the
content
summary
must
be
there
for
getting
eligibility
and
top
20
winners
will
get
some
awesome
sex
that
you
can
check
on
I,
think
website
itself.
B
Yeah,
so
this
is
another
Ambassador
school
board,
so
I'm
showing
this,
because
this
is
also
the
program
ends
and
we
are
having
this
top
10
ambassadors
from
our
past.
I
can
say
past
two
or
three
weeks,
okay,
so
these
are
the
people
who
just
helped
us
a
lot
to
get
some
good
reach
in
our
program
and
lots
of
things
are
there
so
yeah?
This
is
the
program
closed,
and
this
is
the
final
list.
So,
if
you
have
queries
like
I
have
I
am
in
the
combustion
program
of
What's
Done
between.
C
C
Okay,
so
Ricardo
is
an
Enterprise
evangelist
at
Amazon
web
services.
He
works
with
Enterprise,
Executives
and
Technology
leaders
to
help
them
adopt
the
cloud
and
transform
their
businesses.
He
has
over
20
years
of
experience,
leading
emerging
technology
and
Innovation
programs,
and
he
has
been
working
with
Cloud
technology
since
2008,
so
he
is
quite
experienced
and
he
is
literally
a
great
great
speaker
and
he
is
passionate
about
cloud
Innovation
and
open
source
and
is
excited
about
how
Cloud
accelerates
and
amplifies
customers
ability
to
innovate.
So
over
to
you,
Ricardo.
E
Now
I
am
been
working
with
open
source
for
over
20
years
and
I
want
to
share
some
of
the
things
that
I've
learned
in
that
time,
helping
Builders
as
well
as
businesses
to
develop
better
software
and
get
better
outcomes
through
open
source.
But
first
I
want
to
talk
a
little
bit
about
what
actually
is
open
source,
and
why
does
it
matter
now?
E
This,
in
turn
led
to
the
creation
of
the
open
source
initiative
in
1998..
Now
open
source
was
a
shift
away
from
the
free
software
Foundation
to
a
more
commercial
and
business
friendly
way
of
applying
the
same
principles.
They
are
similar,
they're,
not
they're,
not
the
same.
The
free
software
Foundation
used
the
Four
Freedoms
as
its
litmus
test
as
to
what
free
software
should
look
like,
which
is
what
you
can
see
on
the
left
hand
side
here,
The
OSI
created
the
open
source
definition
or
OSD
for
short
and
there's
a
set
of
10
requirements.
E
Where
is
open
source
heading
and
what
does
the
future
hold
now
before?
I
answers
I'm
going
to
come
back
to
this
later,
but
first
I
want
to
talk
a
little
bit
about
why
Builders
love
open
source
now
as
a
future
or
current
Builder.
You
might
be
asking
yourself:
why
should
I
take
notice
of
Open
Source?
It
turns
out.
There
are
many
reasons
why
developers
Love
Open
Source
I'm
not
going
to
cover
all
of
the
ones
here,
but
let
me
just
look
at
a
few
of
them
now.
E
For
some,
it
might
be
having
the
freedom
and
space
to
experiment
with
new
ideas
for
others
it
might
be
scratching
the
itch.
That's
been
the
seed
of
many
new
open
source
projects,
or
it
could
be
that
you
want
to
engage
with
open
source
projects
that
provide
you
with
a
much
more
broad
range
of
an
exposure
to
ideas
and
Technologies
than
you
normally
would
have.
Access
to
open
source
gives
developers
a
way
to
express
their
creativity
whilst
they're
doing
their
work
and
it
eliminates
obstacles
as
well
as
improving
their
productivity.
E
E
One
of
the
key
motivations
when
working
in
open
source
was
that
it
allowed
that
group
of
developers
to
work
at
their
Peak
experiencing
what
he
termed
flow
now
flow
is
a
combination
of
these
three
things:
the
ability
to
have
a
purpose
in
the
work
you're
doing
the
ability
to
work
towards
Mastery
and
expertise,
as
well
as
being
able
to
do
that
under
terms
that
you
are
in
control
of
and
have
autonomy
over
and
when
working
in
open
source
projects.
What
he
found
was
these
developers
were
able
to
achieve
these
three
things
and
work
towards
flow.
E
Now,
according
to
this
Linux
Foundation
survey,
back
in
2019,
62
of
the
developers
said
they
would
be
more
successful
if
they
could
allocate
paid
work
time
to
contribute
to
open
source
projects.
So
it's
clear
that
developers
recognize
the
value
of
participation
and
contributions
to
open
source
project
projects.
E
E
But
it's
also
interesting
to
see
and
read
other
surveys
as
well
that
show
that
Builders
love
to
contribute
to
open
source
projects
for
a
number
of
different
reasons,
both
Technical
and
non-technical.
In
fact,
that
increase
of
soft
skills
is
really
important,
as
you
want
to
collaborate
with
others,
both
in
open
source
or
in
your
commercial.
Day-To-Day
work
and
will
help
you
be
successful
within
your
career.
E
E
The
first
is
that,
whether
you
realize
it
or
not,
you
are
consuming
open
source
software.
This
was
a
recent
report
from
blackduck
who
audited
code
from
Enterprises
and
found
that
96
of
that
code
contained
open
source
components
now,
I've
seen
more
recent
reports
from
black
dark
synopsis
and
red
hat
that
all
point
to
similar
figures
or
impact
higher,
and
they
all
point
to
the
fact
that
you
cannot
Escape
open
source
combined
with
that
software
is
at
the
center
of
digital
transformation,
with
many
Enterprises
now
growing
their
technical
competencies
as
well
as
resources.
E
E
Now,
of
course,
if
you
want
to
be
good
at
software,
you
have
to
be
good
at
open
source
and,
according
to
a
recent
report
from
Forrester
Consulting
90
95
of
those
companies
that
were
surveyed
who
possessed
a
good
open
source
strategy,
either
met
or
exceeded
their
businesses.
Business
expectations
so
open
source
increases
Innovation
by
accelerating
the
adoption
of
three
in
three
ways
by
adopting
the
emerging
emerging
Technologies
by
lowering
the
cost
of
experimentation,
as
well
as
removing
undifferentiated
heavy
lifting.
E
Open
source
lowers
the
cost
of
experimentation,
as
you
can
simply
download
the
software
providing
Builders
with
a
really
easy
and
quick
way
to
try
explore
experiment
without
having
to
pay
any
software
licenses
and
open
source
projects
contain
the
building
blocks
on
which
you
can
use
to
build
your
own
products
and
services,
and
these
are
critical
when
you're
looking
to
accelerate
Innovation
activities
such
as
prototyping.
Could
you
imagine,
for
example,
today
trying
to
build
a
new
approach
type
without
your
favorite
open
source
building
blocks?
E
So
back
to
the
future
of
Open
Source
as
Builders
and
businesses
grow
in
their
use
of
an
adoption
of
Open
Source
Technologies.
One
of
the
challenges
that
I
began
to
see
many
years
ago
was
the
more
time
spent
managing
the
running
and
operations
of
Open
Source,
as
well
as
seeing
challenges
in
the
support
of
the
increasing
number
of
those
open
source
Technologies.
They
want
to
use,
and
the
conversation
quickly
turned
from
I
want
to
run
this
open
source
technology
to
I
want
someone
else
to
run
it
and
operate
it
for
me.
E
Reviewing
the
code
and
all
bugs
are
shallow,
and
in
many
cases
this
can
improve
the
quality,
as
well
as
the
security
of
the
open
source
software.
When
we
created
s2n,
which
is
our
Apache
2,
licensed
TLS
implementation,
we
found
it
beneficial
that
we
were
able
to
put
the
code
out
there
and
get
much
more
feedback
from
developers
who
were
able
to
review
that
code
and
provide
feedback
on
what
they
saw.
E
E
Sometimes
those
contributions
lead
to
new
projects
that
we
open
source,
such
as
the
ones
that
I'm
sharing
here,
including
the
ones
I've
mentioned,
such
as
firecracker
and
bottle
rocket
and
new
ones
that
are
coming
out,
such
as
Babelfish
as
well
now
across
both
consuming
and
contributing
to
open
source.
This
Force
multiplier
effect
is
the
tide
that
lifts
all
boats
and
allows
open
source
to
win
for
everyone,
combining
the
best
open
source
and
Cloud
to
accelerate
business
outcomes
enabling
Innovation
and
powering
digital
threat
transformation.
E
Now,
events
such
as
OS
Tech
are
only
possible
thanks
to
the
general
support
from
the
sponsors,
so
I
want
to
thank
the
Platinum
sponsors,
red
hat,
reskill
and
cncf,
so
make
sure
you
check
out
their
sessions
tomorrow
on
openshift
and
voyaging
the
cloud
native
Seas.
Thank
you
to
all
the
other
sports
as
well,
so
make
sure
that
you
spend
time
with
them
too.
E
B
B
B
Regularly
speaks
at
events
and
meetups,
and
yesterday
also
I
was
checking
on
thread
on
Twitter
and
just
tactic
like
some
of
positive.
He
tweeted
Michael's,
who
sent
his
such
a
great
speaker,
so
you
all
will
be
enjoying
his
talk
and
Michael
hustle
weekly
technology,
coffee
chat,
called
everyone
can
contribute
Cafe.
Okay,
so
you
can
just
join
them
on
everyone
can
contribute.com
and
if
you
want
to
contribute,
I
can
see
open,
suits
or
any
of
the
cicd.
You
want
to
learn
something.
B
E
F
I'm
gonna
like
I,
want
to
share
today
my
story:
around
How
I,
Learned
infrastructures
code,
the
first
steps
as
a
developer
coming
to
Cloud
native
deployments
development
and
how
this
could
work
in
five
minutes.
So
this
is
like
a
story
of
my
own.
In
the
past
10
years,
I
joined
gitlab
in
March
2020,
before
that
I
was
involved
in
monitoring,
open
source
development
and
basically
yeah
I
had
had
lots
of
fun
learning
things
and
also
like
trying
new
things
and
breaking
things
and
other
things
stuff.
F
Dns
me
he
so
this
is
like
Michi
is
the
lovely
form
of
Martin
in
German
reach
out
to
me
on
Twitter,
on
LinkedIn,
on
gitlab
or
on
gitlab.com,
or
you
can
also
find
me
on
polywork
recently,
but
yeah,
let's
dive
into
the
magic
stuff
and
in
basically
the
interesting
stuff,
okay,
just
a
little
agenda,
we
will
start
out
with
a
developer's
tale
of
myself
we're
gonna
dive
into
like
learning
packaging.
What
is
that?
What
is
infrastructure
management?
What
are
containers
later
on
figuring
out?
Oh
the
FCI
CD.
F
My
Learning
Journey
started
some
years
ago
and
one
of
the
like
most
interesting
things,
I
also
shared
with
Andy
Gardner
in
the
Pure
Performance
podcast
this
week
was
to
a
boost
from
33
to
66
megahertz.
So
we
had
this
button
on
the
the
computer.
The
PCD
hardware
and
I
was
fascinated.
I
didn't
know
what
what
this
is
and
I
wanted
to
learn
how
this
works.
So
I
started
hardness
of
this
systems,
engineering
and
then
I
figured
out.
Oh
circuit
engineering
is
not
my
profession.
F
I'm
more
in
the
software
side
of
life
learned
about
NFC,
then
multimedia
streaming
on
Linux
later
on.
Dns
development
had
no
idea
about
DNS
in
2009,
but
it
got
a
hold
on
my
nickname,
so
it
stayed
for
a
while
I
might
be
naming
myself
git.
Let
me
get
me
here,
but
I'm
sticking
with
DNS
machine,
so
yeah
this
is
basically
I
was
curious
to
learn
things.
F
At
some
point
travel
the
world
again
now
when,
when
the
developer's
tale
starts,
we
are
starting
out
to
build
our
own
Lego
models
or
Lego
bricks,
and
this
is
one
of
the
first
things
I
experienced
when
I
was
writing
my
first
code
some
several
years
ago
or
like
10
or
15
years
ago,
you're
writing
the
first
web
application
you're
creating
a
top
or
sip
archive,
and
then
you
send
it
over
to
someone
else
like
maybe
email
or
something
else
and
ask
them
hey.
Can
you
please
install
it?
F
F
The
other
feedback
you
might
be
getting
is
well.
The
protection
environment
is
different
to
the
developer
environment.
We
are
calling
making
solid
fails.
Some
module
is
not
installed.
That's
interesting!
What's
that,
because
it
works
on
my
machine,
so
maybe
at
dependency
management
add
something
which
checks
that
the
packages
are
installed
like
similar
to
configure
Auto
conf,
totally
new
language
to
learn
as
a
developer,
not
much
fun
to
learn
to
be
honest,
but
still
it
provided
more
ways
to
check
the
dependencies
I
installed
now
coming
from
the
dependency
checks.
F
Actually
the
request
was
well.
We
want
to
have
packages,
so
please
go
ahead
and
our
customer
wants
RPM
packages.
Okay,
as
a
developer,
I'm.
Looking
at
it
and
say:
hey
the
spec
file,
this
looks
easy
doing
things
and
stuff
like
that,
but
then
you
need
to
make,
for
example,
conditions
with
macros,
and
this
is
like
where
things
are
starting
to
break
and
it's
not
not
easy
anymore.
F
The
other
side
is,
after
you
manage
to
build
a
build
RPM
packages
and
other
customers
as
well.
We
have
DB
and
we
have
Ubuntu.
We
have
other
the
robots
of
that.
We
want
to
have
Debian
packages
now
as
a
developer,
you're
sitting
in
front
of
him
saying:
okay,
this
is
a
totally
new
tool
chain
work.
There
are
so
many
different
things
to
understand
and
my
head
is
exploding.
F
I
also
need
to
think
about
dependency
management
again
and
actually
I
just
want
to
escape,
or
just
one
I'm
not
really
sure
about
that.
But
still
it's
it's
a
huge
Learning
Journey.
The
other
thing
is
we
wanted
to
have
a
build
server,
so
basically
CI
CD
or
something
which
just
executes
and
automates
stuff.
F
So
we
want
that
so
Ops
Ops,
Team
back
then
said
well,
maintain
it
yourself:
okay,
let's
install
Jenkins
and
start
or
Hudson
Ci,
or
was
it
back
then
actually
and
let's
run
it
and
then
the
security
team
or
like
Ops
doing
security
was
coming
back
after
a
while
saying,
hey:
why
does
the
demon
run
this
route?
Why
is
the
API
exposed
without
password?
Why
is
everything
broken
now,
because
an
attacker
found
the
way
in
our
Network
so
well
developers?
Maintaining
tools
might
not
be
the
best
idea.
F
In
that
case,
the
other
thing
is
what
we
needed
to
learn
over
time.
Well,
we
need
not
only
need
to
maintain
one
single
package.
We
need
to
maintain
architectures.
We
need
to
learn
how
different
how
different
Behavior
actually
happens
on
X
for
for
the
64
or
x68
or
specific,
on
embedded
Hardware
with
arm
with
different
platforms.
F
Then
we
had
a
bunch
of
distributions
and
you're
dependent
on
Old
build
tool
chains.
You
have
ground
dependencies.
You
cannot
really
use
the
modern
design
patterns.
You
want
to
do
like
C,
plus,
plus
17
or
C,
plus
plus
14,
in
that
regard,
or
a
specific
other
modern
JavaScript,
because
you're
still
supporting
Internet
Explorer
6.
What
was
it
five
back?
Then?
I
don't
know
still.
Maybe
you
want
to
go
back
to
supporting
just
one
platform,
because
you
want
to
stay
productive
and
efficient,
but
yeah
it
just
turned
out
to
be
well.
F
We
can
automate
things.
We
also
need
to
look
into
not
only
like
packaging
all
this
stuff,
but
we
need
to
manage
the
infrastructure
or
config
management
how
it
was
called
before
so
the
Ops
Team
actually
deploys
the
software,
so
there
are
different
tools
out
there,
and
these
are
just
a
few
examples
of
like
the
puppet
agent.
Then
she
was
the
latest
version
installed.
Ansible
does
kind
of
the
same.
F
F
Ops
also
needs
to
do
more
in
that,
and
this
is
a
certain
way.
Dev
also
needs
to
understand,
like
you
can
manage
the
base
system
the
dependencies,
but
you
can
also
like
manage
users
which
need
to
exist
that
the
specific
Frameworks
you
need
the
container
runtimes
the
container
cluster
orchestrators.
You
need
to
take
care
about
backups
because
well,
you
need
to
take
them,
but
you
also
need
to
test
them,
and
monitoring
is
also
a
big
part.
F
If
you
just
leave
it
out
of
your
deployment,
you
won't
see
anything,
you
won't
get
any
alerts,
so
it
needs
to
be
done
and
also
needs
to
be
done
in
the
workflow
in
the
process
of
provisioning.
Something
now
Ops
comes
back
to
the
developers,
because
the
developer
wants
to
deploy
a
new
application,
and
then
there
are
some
requirements
to
Define
and
it's
up
to
the
developer.
To
say:
hey,
please
specify
the
database
size
or
the
the
Ops
people's
Ops
Team
says:
I'm.
Sorry,
I
have
no
idea.
F
F
So
yeah,
it's
it's
a
lot
of
things
being
asked
for
now
when
we
move
on
into
like
Ops
wants
things
from
developers.
Well,
when
when
it
comes
to
monitoring,
you
need
to
have
certain
thresholds,
so
you
get
an
alarm
when
something
is
like
expected
to
be
breaking
or
already
broken.
Like
CPU
usage
goes
up,
Beyond
disk
is
full
and
the
thing
is
you
when
you're
collecting
metrics,
you
cannot
store
them
endlessly.
You
are
in
infinite.
You
need
to
like
do
think
about
retention
so
like
when
you
store
everything.
F
The
back
end
will
be
full
at
some
point.
The
other
thing
is,
you
cannot
cover
everything
with
monitoring
with
things
which
are
already
there.
So
Ops
might
say:
hey
I
want
to
see
more
runtime
metrics
of
the
application,
so
the
developers
need
to
learn
Prometheus.
They
need
to
learn
about
exporters.
They
need
to
learn
how
slash
metrics
as
an
HTTP
endpoint
works,
and
what
is
that?
No
idea?
F
On
the
other
side,
you
might
be
asked
for
hey
this.
This
is
amazing
that
we
have
a
log
when
the
narrow
occurs,
but
this
is
like
the
long
stack
Trace.
We
need
to
scroll
10
pages,
but
we
don't
really
see
the
error
in
in
the
first
second,
so
thinking
about
structured
logging
and
I
was
attending
a
talk
from
from
Nicola
from
Kell
a
while
ago,
where
I
actually
learned
about
the
the
benefits
about
structured
logging
like
making
it
processible
for
machines,
you
can
see
it
or
correlate
it
with
metrics.
F
You
can
immediately
see
what's
going
on
on
the
machine,
because
potentially
the
machine
is
dead
and
SSH
doesn't
work.
So
you
can't
see
the
logs
with
tail
Dash
f,
and
this
goes
on
like
distributed,
tracing
comes
to
comes
to
play.
So
what
is
that?
Is
that
different
to
an
event?
Is
it
different
to
a
log,
oh
spans
and
traces,
to
immediately
see
different
steps
in
an
HTTP
request
into
the
back
end
into
the
postgresql
database?
Seeing
that
the
red
is
caching
mechanism
is
blocking
and
even
more
in
the
microservice
world.
F
Now,
when
something
breaks
you
need
to
call
someone
or
someone
needs
to
be
on
call
so
like
some
people
are
saying
that
devops
means
steps
on
calls
actually
well
yeah.
We
should
we
should
work
together
as
a
team,
but
it
doesn't
mean
the
devs
need
to
be
on
caller,
because
the
Ops
team
will
get
paged
anyways.
So
there
is
a
certain
control
mechanism.
Oftentimes.
F
You
need
to
Define
who
theorize
the
directly
responsible
individual
who's
on
who's
on
call,
but
it's
actually
like
you're
training.
Your
team
card
Trend
handle
it
together
in
a
specific
sense.
Now,
when
the
journey
continues
at
some
point,
someone
said
Hey
There
Will
stock
are:
there
are
containers.
There
is
cloud
native.
F
There
are
so
many
new
things
to
learn,
but
what
is
that
now,
when
you
started
or
when
I
started
out
learning
Docker
and
seeing
the
benefit
from
my
own
systems,
like
a
small
sandboxy
development
environment,
for
instance,
or
from
a
deployment
angle,
throwing
away
all
the
Debian
and
RPM
stuff
I
needed
to
learn
and
just
using
a
container?
But
what
are
the
best
practices?
So
I
cannot
remember
the
difference
between
ADD
and
copy
in
a
Docker
file.
I
always
need
to
Google
it
and
I
I
gave
the
same
talk
at
dockercon
and
was
like
okay.
F
It's
actually
not
me,
not
just
me
not
not
being
able
to
remember
the
thing
is
there
are
so
many
other
things
to
learn
like
there
is
a
Docker
file.
It
has
a
specific
syntax.
Then
you
have
the
docker
compose
which
brings
up
an
entire
stack
of
containers.
This
is
a
new
Yammer
format
to
learn.
I've
seen
a
builder
image
in
specific
use,
cases
and
I
was
wondering
what
is
that
and
then
I
started
to
dig
and
learn
and
try
and
yeah
at
some
point,
I
figured.
Oh,
this
is
actually
a
best
practice.
F
The
other
thing
I
I
was
reading
I
think
it
was
2015
or
something
the
book
about
around
Docker
and
containers
and
stuff
like
that,
and
at
first
glance,
I
thought
well.
I
can
use
a
container
just
like
a
virtual
machine
using
a
supervisor
CTL
instead
of
systemd,
but
instead
you
should
be
just
running
one
service
inside
a
container,
so
you
can,
you
can
make
many
mistakes
and
I
obviously
did
the
other
thing.
Is
there
there
are
no
or
in
most
recent
times
this
got
better.
F
F
F
And
this
is
an
end
of
Adventure
on
its
own,
like
okay,
we
have
a
Docker
file.
How
can
we
build
that?
There
is
a
way
of
doing
Docker
in
Docker,
which
is
some
magic,
Unix,
socket
mapping
inside
and
then
you're
building
a
an
image
inside
the
container
and
pushing
it
to
the
outside
I
think
this
took
me
some
months
to
understand
on
the
horizon
or
there's
a
new
API
with
conico.
F
This
doesn't
need
root
demons
on
the
side,
Portman
is
being
developed,
but
it
to
my
knowledge.
It's
not
a
hundred
percent
compatible.
Yet
with
all
the
docker
features
available,
and
the
other
thing
you
need
to
think
about,
is
it's
fine
to
build
everything
and
leave
it
there,
like
you,
you're,
creating
git
repositories
on
GitHub
or
gitlab,
and
you
don't
clean
it
up
because
yeah
it's
just
there,
nobody
cares.
Basically,
the
thing
is
like
creating
Docker
images
is
also
consuming
like
disk
space
in
that
regarded
at
a
certain
points.
F
This
Quest
is
running
out
and
the
typical
DF
Dash
H
checks,
don't
work
so
disk
usage
just
because
Dockers,
using
a
different
virtual
file
system
and
and
other
things.
In
the
background,
so
you
really
need
to
analyze,
what's
consuming
disk
space
and
not
just
doing
an
rm-rf,
but
instead
like
Docker
image
prune
something
I
need
to
Google
again
just
to
ensure
that
you
don't
you're,
not
breaking
anything
and
freeing
up
this
space
for
future
operations,
now
lots
of
things
to
learn,
but
it
doesn't
stop
here.
F
The
other
thing
I
I
figured
out
was
okay.
Now
that
I
have
a
Docker
image,
I've
built
it
locally,
it's
amazing,
I've
tagged
it,
but
actually
I
want
to
make
it
available
for
others,
and
there
is
Docker
Hub,
obviously,
but
what,
if
I'm
in
a
local
network
in
a
DM
set
or
something
else,
I
want
to
just
have
it?
F
The
best
thing
is-
and
this
is
an
example
from
from
the
gitlab
configuration
you're
using
a
built-in
template
and
it
does
all
the
magic
for
you,
but
getting
there
is
a
learning
process
yeah
and
like
relying
on
the
specific
back-end
demons
who
are
required
for
doing
so.
Like
talk
and
talk,
and
that
example
yep
lots
of
things
to
learn
now:
automating
everything
and
Diving
a
little
more
into
what
is
continuous
integration,
continuous
delivery
or
continuous
deployment.
F
At
first
glance,
we
started
started
out.
Well,
let's
start
with
CI.
Let's
start
with
the
simple
things
and
at
a
certain
point,
my
Branch
commit
history
was
a
little
longer
because
well
I
couldn't
figure
out
how
to
create
the
best
or
the
correct
Yammer,
and
this
was
just
trial
and
error
to
figure
out
how
the
format
works.
This
got
better
over
time
with
now
pipeline
editors
and
linting,
and
automatic
checks
and
stuff
like
that.
F
But
when
I
started
out
with
that,
I
think
that
screenshot
is
two
years
old
or
something
it's
a
lot.
It
is
a
lot
of
lots
of
trial
and
error,
and
everyone
on
the
team
tries
that
now
that
you
have
your
first
success,
it's
amazing,
but
the
pipeline
is
taking
10
minutes
to
run.
So
you
might
be
credit
or
asked
to
optimize
that,
especially
because
every
time
a
pipeline
is
running
and
you're
trading
an
MR
or
a
PR,
you
need
to
wait
for
that.
So
it's
like
you
cannot
do
anything.
F
You
need
to
focus
on
other
things
or
use
XKCD
303
as
an
excuse
and
yeah.
It's
it's
on
the
way
to
like
learning
how
to
optimize
it
and
at
a
certain
point
developers
become
Ops
because
well,
please
fix
the
CI
CD
pipelines,
because
you
are
maintaining
the
configuration
you're
responsible
for
that,
and
why
should
I
do
that?
Yeah?
It
consumes
so
many
resources
and
then
like
how
yeah,
maybe
the
documentation
is
there,
someone
wrote
it
asked
the
community
and
do
something
about
it.
F
So
it's
like
who
again
learning
learning,
learning
and
especially
diving
into
areas
you
might
never
have
been
before
now.
Well,
this
is
like
this
is
only
a
summary
of
what
you
can
do
or
can
start
with
to
to
identify
bottlenecks
so
like
identifying
the
pipeline
duration,
seeing
the
critical
path
like
the
minimum
and
the
maximum
duration
inside
the
CI
CD
pipeline.
Looking
at
stages
and
jobs,
dependencies,
Sun,
currants
or
async
runs
in
that
specific
regard,
identifying
patterns
of
failed
jobs
from
flaky
tests.
F
So
one
time
at
the
day,
the
tests
are
running
good
the
other
day.
It's
like
something
is
broken,
and
maybe
you
can
correlate
that
to
specific
Network,
latency
events
or
something
else.
It's
like
influencing
that
similar
to
Containers
dependencies
being
installed
all
the
time
packages
and
stuff,
the
End
security
and
firewalls
the
network
and
all
the
things
it's
like
you.
F
For
for
for
operations,
like
am
I
using
a
container
orchestrator,
should
it
be
like
Docker
compose
swarm,
kubernetes
or
on
the
other
side,
and
by
provisioning
in
AWS
ec2
instance,
with
using
infrastructure
as
codes,
something
like
terraform
and
server
Automation,
and
the
last
thing
is
like
it's
not
only
installing
the
application.
One
time
you
need
to
ensure
that
they
are
running,
you
need
to
ensure
that
the
boxes
or
the
the
application
and
the
operating
systems
are
upgraded.
And
potentially
you
need
to
do
some
hot
fixing
and
security
packs.
F
That's
all
you
need
to
keep
in
mind
for,
for
that,
and
at
a
certain
point
we
are
back
at
hey.
Developers
are
now
packages
again
like
Cloud
packages,
because
well
the
docker
image
best
practices
are
still
there.
We
have
shifted
from
okay,
it's
not
RPM,
it's
not
the
dbn,
it's
not
a
turbo
anymore.
It's
a
Docker
image,
but
still
we
need
to
think
about
like
Builder
images,
yeah
one
thing
and
the
other
thing
is
like
keeping
the
images
small.
F
The
other
thing
is
like
okay,
there's
kubernetes
do
I
need
that
if
I
need
it,
how
can
I
learn
it
and,
to
be
honest,
it
took
me
quite
a
long
time
to
to
Really
find
the
use
case.
Find
the
way
of
saying
hey.
I
have
understand,
understood
it.
I
I
know
how
to
write,
Yammer
configuration
and
not
just
doing
copy
paste
from
some
specific
documentation
or
tutorial
I
found
on
the
internet.
Next
up
or
keyboard
is
in
health
chart.
How
can
I
install
applications
and
software
inside
kubernetes?
What
is
an
operator?
F
F
The
other
thing
is
cloud
native,
build
packs,
which
is
a
similar,
similar
idea
sitting
in
front
of
it
and
say:
okay,
this
is
useful,
but
how
can
I
make
use
of
it?
How
does
it
solve
my
problem?
Still
it's
a
huge
Learning
Journey
well,
and
then
there
is
something
like
Auto
devops,
so
like
thinking
around
hey,
maybe
I
can
optimize
all
of
that
saying:
okay,
I'm
I'm
committing
it
and
it
gets
deployed
to
production.
So
you
don't
need
an
extra
configuration
it
does.
F
F
So
this
is
one
of
the
ideas
you
could
potentially
be
looking
at,
but
the
thing
is
I
need.
I
need
to
understand
the
black
magic
because,
especially
when
it
breaks
I
need
to
fix
it
now.
Well,
after
all,
as
a
developer,
I
just
want
to
deploy
my
web
app
I,
don't
really
care
whether
it's
q
a
is
or
it's
ec2
or
something
else.
F
This
is
like
tackling
all
together
and
some
recursive
infinite
loops
and
then
stuff
like
that.
The
thing
is:
let's
go
back
to
saying:
hey:
what
is
the
goal?
What
what
are
the
sides?
What
is
the
side
from
the
developer?
Obviously
we
want
to
deploy
fast,
so
we
we
want
some
something
to
build
it
automatically.
Tests
should
be
there
tracing
scanning
security
scanning.
F
This
should
be
like
somehow.
Automated
deployment
should
be
automated
and
at
a
certain
point,
when
security
fixes
are
done
by
the
penderbot
or
I
think
there's.
There
are
different
tools
out
there,
which
can
be
automated,
maybe
adding
Auto
merge,
but
in
this
in
a
sense
of
combining
it
with
security
scanning,
so
that
no
one
injects
some
crypto
mining
stuff
or
whatever
into
your
pipelines
or
your
code
now
getting.
F
There
is
again
in
a
Learning
Journey
and
then
defining
defining
the
resources
analyzing
what's
needed,
like
okay,
we
need
to
provision
Cloud
resources
for
that
becoming
Cloud
native.
Maybe
we
will
be
using
virtual
machines.
We
will
be
using
services
so
using
not
installing
a
postgresql
database,
but
just
using
or
Aurora
I
hope.
I
pronounced
it
correctly,
as
as
a
database
service
in
in
AWS,
for
instance,
or
for
example,
I
could
be
using
a
redis
cluster.
To
be
honest,
I
have
no
idea
how
to
run
it.
F
Okay,
we
did
documentation
and
and
make
it
work,
but
still
I
just
want
to
consume
it.
I
want
to
provision
it
and
I
also
want
to
use
object.
Storage
using
S3
or
something
compatible
in
that
way,
the
other
thing
is
or
the
other
way
I
need
to
think
about.
This
Cloud
providers
have
an
API
I
need
to
interact
with
the
API
AI.
F
Hopefully
there
is
an
SDK
in
my
preferred
language,
for
Ruby,
for
golang,
for
whatever
and
I
also
need
to
think
about
user
management,
and
this
is
something
where
it's
a
mix
of
Dev
and
Ops,
defining
the
requirements
and
then
actually
making
use
of
that.
Now,
when
we're
developing
an
application,
we
might
be
needing
certain
additional
things
like
if,
when
it's
a
web
application,
we
might
need
a
persistent
state
in
the
back
end.
So
it's
not
just
in
memory.
F
Speaking
of
that
being
up
to
date,
TLS
certificates
with
let's
encrypt
on
necessary,
are
important.
This
is
something
which
you
can
also
automate
with
specific
Acme
challenges
and
whatnot
still.
It
needs
to
be
automated
and
if
there
is
the
need
for
specific
TLS
endpoint
with
an
nginx
prox
up
front,
for
instance,
just
rooting
into
the
containers.
This
is
something
which
also
needs
to
be
done.
So
this
isn't
a
requirement
to
be
known
about,
but
hopefully
we
can
automate
that
the
other
thing
is
like
for
automation.
F
We
can
really
use
what
we've
learned
earlier
about
infrastructure's
code,
so
we're
using,
for
example,
ansible
and
terraform
to
provision
the
cloud
resources
which
are
needed.
It's
like
infrastructure,
as
cultures
means
you're,
describing
what
you
want
to
see
or
want
what
you
want
to
have
in
a
code
Cody
way,
and
it's
not
likely
kibonti
or
like
clickbonder,
doesn't
work
in
in
English
just
clicking
in
an
interface
and
it's
colorful.
You
really
automate
that.
F
So
the
other
side
is
like
inventory,
seeing
what's
actually
provisioned
what
is
generating
compute
costs
and
and
whatnot
with
a
deeper
integration
into
the
git
backends
or
the
git
repositories.
You
can,
for
example,
use
gitlab
as
a
terraform,
State
backend.
You
don't
need
to
have
any
specific
other
things
around
it.
It's
just
there.
F
You
can
make
use
of
it
and
you
can
use
the
ICD
variables
to
actually
create
pipelines
which
do
the
deployment
and
Paramount
parametrize
that
so
like
instrumenting
the
pipeline
to
behave
as
a
release
power
plan
as
a
debug
pipeline
as
a
staging
Pipeline,
and
what
not
okay,
long
story
going
or
taking
again
what
what
is
already
there.
Maybe
we
still
want
to
use
Auto
build
no
Docker
file.
We
don't
want
to
learn
about
Docker
in
that
regard.
I
want
to
use
Auto
build
and
the
hero
Kush
Builder
the
cloud
native,
build
packs.
F
It
gets
deployed
at
a
certain
point,
but
before
it
hits
the
deployment,
we
need
to
ensure
certain
things.
So
the
deployed
app
needs
to
pass
quality
Gates
like
code
quality,
is
something
happening
and
decreasing
the
code
quality
is
the
algorithm
getting
more
complex,
the
architectural
complex
these
checks,
which
can
be
automated
based
on
open
source
tools.
F
The
other
thing
is
like
service
level
objectives
like
my
git
commit
changed
the
performance
or
decrease
the
performance,
because
it
consumes
10
times
more
memory,
and
this
is
something
you
can
actually
detect
in
a
staging
environment
being
run
and
the
rip
the
SLR
reporting,
basically,
for
example,
with
Captain,
as
shown
in
the
screenshot
Can,
can
detect
that
and
can
prevent
that.
Actually,
this
is
being
pushed
into
production.
F
Having
that
in
CI
CD
in
the
deployment
also
provides
you
a
single
source
of
Truth,
so
you
know
where
the
current
state
is.
You
know
where
the
terraform
state
is,
for
example,
you
know
how
the
infrastructure's
code
gets
provisioned
next
to
the
code
itself,
so
you
can
make
use
of
in
your
infrastructures
code
CI,
CD
templates
and
also
we
use
all
the
best
practices.
Others
already
have
learned
and
worked
together
on
a
common
solution
and
not
just
Reinventing
the
wheel
all
the
time.
F
The
other
thing
is
as
a
developer,
it's
similar
to
kubernetes
or
to
the
deployment
I,
don't
really
care
how
terraform
works,
I
want
it
to
work,
and
if
it's
there,
I
probably
won't
be
touching
it
or
just
reusing
a
template.
What's
already
there
and
infrastructures
code
can
be
used
to
provision
the
virtual
machines
or
the
resources
and
later
on,
that's
the
inventory.
If
I
don't
want
to
use
a
container
cluster
like
kubernetes
I,
don't
use
it.
That's
it
and
optionally.
F
This
is
one
of
one
of
our
Visions,
which
we
have,
and
it
also
shows
the
path
forward
on
saying:
hey
as
a
developer.
I
want
my
app
being
deployed
in
five
minutes:
I,
don't
care
how
it
works
in
the
background
it
does
it
magically
for
me,
and
hopefully
it
does
it
right
and
it
gets
monitored
at
everything
else,
which
is
best
practices
works
for
us
and
we
can
focus
on
the
development
on
the
things
which
we
are
good
at
looking
into,
like
optimizing,
the
architecture
optimizing,
the
performance,
learning
about
monitoring,
metrics
and
whatnot.
F
Now
there
is
a
slight
problem
with
the
adventure
of
the
Tyranny
over
the
past
years,
like
burnout
is
real
and
developers
need
to
learn
so
so
many
things
that
is
really
like.
You
can't
just
replace
a
team
of
four
or
five
people
or
three
people
with
one
devops
engineer
that
doesn't
work
over
time.
You
will
be
burning
out
teams
or
single
persons,
so,
like
you
need
to
really
focus
on
what
you
want
to
learn
and
what
you
want
to
do.
F
So
when,
when
you
start
out
your
first
Learning
Journey
of
your
first
devops,
Adventure,
I,
would
say,
I
would
recommend
to
focus
on
building
bricks.
So
you
start
learning
in
a
different
environment.
You
leave
everything
which
is
like
all
the
platforms
are
tested
that
that
this
is
like
bound
to
cicd
workflows.
This
is
something
which
runs
when
you
push
the
code
automatically,
but
it
doesn't
need
to
be
on
on
your
on
your
focus
right
now.
F
The
other
thing
is,
you
want
to
use
infrastructure
as
code
for
automation,
to
deploy
faster
at
a
certain
point
in
your
Learning
Journey,
but
not
in
the
beginning.
Don't
feel
overwhelmed
by
that.
The
other
thing
is
learning
alone
is
oftentimes
hard
and
sitting
in
front
of
an
online
course
is
also
oftentimes
hard.
Try
to
ask
a
friend
or
join
a
community
to
do
peer
programming,
group,
programming,
group,
debugging
and
stuff
yeah.
F
One
of
the
things-
and
this
is
a
blog
post-
I
wrote
this
week
with
Marcel-
is
like
shift
the
development
of
elements
to
the
cloud
so
make
it
easier
to
contribute,
because
local
setups
differ
the
errors
too.
You
can
easily
join
a
team.
You
can
do
a
stream
where
you
develop
something
together
like
we
started
out
learning
rust
a
while
ago,
just
in
the
cloud
in
the
browser,
because
the
environment
is
already
there
and
it
actually
Visual
Studio
code
in
your
browser.
F
The
other
thing
is
and
I
I
need
to
speed
up
and
hopefully
I
nearly
at
the
end.
It's
lots
of
things
to
learn.
Obviously,
but
I
also
want
to
give
you
some
tips
and
tricks
what's
what's
coming
and
what
you
can
prevent
of,
like
you,
don't
need
to
know
everything.
I
don't
even
know
everything
in
that
regard.
The
thing
is:
if
we
jump
into
like
understanding,
monitoring
and
see
the
value
of
metrics
logs
and
traces
and
specific,
like
is
it
monitoring?
Is
it
observability
diving
into
the
the
thing
like?
F
Is
it
like
a
bug
in
the
code
or
is
it
the
system
which
got
broken
or
is
it
the
container
cluster
which
has
an
issue
like?
Is
the
load
balancer
Network
routing
broken,
for
instance,
so
this?
This
is
something
you
need
to
keep
in
mind
when
you
look
into
implementing
things
and
adding
more
insights
into
the
application
for
others
now
approaching.
This
is
com
like
a
complex
topic,
and
one
of
the
things
I
would
recommend
is
to
start
simple
use.
F
The
next
thing
is
like
building
a
dashboard
with
this
is
like
a
an
opinion.
Opinionated
recommendation
of
mine
start
with
Prometheus
for
metrics
and
monitoring
start
with
Loki
and
then
promptail
or
fluenti
for
logs
and
combine
that
in
grafana
and
when
an
incident
is
happening.
You
are
clicking
on
the
metrics
board
or
Matrix
panel
on
on
the
spike,
and
you
can
see
the
logs
immediately.
What
what
causes
that
or
what
correlates
to
that
yep!
F
That's
the
thing.
The
other
thing
I
would
recommend
is
don't
do
everything
yourself
because
really
like
it's
too
much
to
learn
too
much
to
do,
even
even
if
in
the
huge
team
you
might
be
seeing
that
not
everything
happens
and
the
backlog
is
filling
up
so
start
start
thinking
about
like
not
that
written
test
makes
sense,
or
maybe
you're
lazy
like
me,
and
you're
not
lacking
to
write
unit
tests
all
the
time.
The
other
thing
is
people
who
might
say
100
discouraged
yeah.
No,
that's
not
possible
there
is.
F
You
cannot
cover
everything
in
your
code.
Sometimes
you
cannot
really
create
a
Mark
or
an
aspect
test
for
that
yeah.
The
other
thing
is
to
think
about
is
like
bringing
in
chaos
and
known
data
into
the
devops
life
cycle
and
saying:
hey.
There
are
security
scandals
out
there
I
can.
We
can
make
use
of
that,
put
it
in
CI,
CD
and
we'll
get
feedback.
We
can
learn
how
this
works.
We
can
act
on
on
the
feedback
based
from
the
security
scan.
F
We
can
add
fast
testing
like
the
unexpected
tests
for
our
apis.
We
can
even
introduce
chaos
engineering
into
into
our
devops
life
cycle
and
say:
hey
we're
using
litmus
chaos,
for
example,
and
then
defining
some
test
cases
and
it
really
better
tests
the
application
in
an
unknown
environment
coming
from
chaos,
engineering.
You
also
need
to
practice
observability
Methods
at
a
point
like
using
the
slos
for
Quality
Gates
thinking
about
not
what
is
there
but
embrace
the
unknown.
F
Anna
is
always
created
the
the
100
days
of
kubernetes
platform,
which
is
a
huge
knowledge,
knowledge
base
for
learning,
cloud,
native,
kubernetes,
monitoring
and
other
things.
This
is
a
great
resource
for
starting
a
Learning
Journey
and
for
live
streams.
There
is
cloudnative.tv
and
I'm,
also
hosting
my
own
weekly.
Everyone
can
contribute
Cafe,
which
is
like
the
the
tech
coffee
chat
for
learning
together
from
it's,
not
it's.
Sometimes
we
touch
gitlab
as
well,
but
we
are
keen
on
learning
things.
So,
like
learning
graphs
learning
this
week,
we
had
civil
cloud
with
k3s
clusters.
F
We
had
Ops
Trace,
which
is
monitoring
security.
So
whatever
is
interesting
for
us.
We
want
to
learn
together
and
we're
live
streaming
that
on
gitlab
as
well
on
unfiltered
so
yeah.
This
is
about
it
and
I
talked
too
much
today.
Thanks
for
your
attention,
keep
let's
keep
in
touch.
Dns
me
here
over
there,
and
now
we
have
a
little
time
for
questions.
I.
Think.
A
B
F
Yeah
the
thing
the
thing
with
like
devops,
defining
it
as
a
workflow,
defining
it
as
a
team
culture,
something
you
need
to
practice
and
you
need
to
like
find
the
right
tools
and
the
processes
for
your
teams
and,
if
you're,
taking,
for
example,
Auto
devops,
how
it's
called
in
gitlab.
It's
it's
already
like
it's
an
opinionated
workflow
in
an
in
crcd
templates
and
and
build
tools
which
you
can
make
use
of,
and
you
don't
need
to
like
learn
it
from
scratch.
F
B
B
Yeah
so
Fox,
the
first
thing
I
want
to
elaborate
before
starting
the
quiz
is
like.
So
this
is
the
quiz
link.
Okay,
we
are
sharing
you
in
the
comments.
Also,
okay,
so
you
can
find
the
quiz
Link
in
the
comment
section
and
what
you
have
to
do.
You
have
to
just
go
and
just
join.
This
particular
I
can
say,
link
okay
and
whenever
you
started
joining,
you
will
get
your
entries
here,
so
you
can
just
find
it
out
whenever
you're
joining
you
will
get
your
entries
here.
B
So
we
have
already
put
the
I
can
say
just
join
it,
okay,
so
what
you
have
to
do?
You
have
to
just
put
the
quiz
Link
in
your
phone
or
I
can
say
in
a
new
tab.
Okay,
so
that's
how
you
have
to
join
it.
So
I
can
see
a
lot
of
people
are
joining
in
that's
awesome
and
we
have
created
the
quiz
according
to
the
session
itself,
if
you
have
focused
in
session,
you
have
learned
offer.
If
you
have
listened
about
iaac,
then
you
must
be
able
to
answer
the
quiz
also.
B
Yeah,
so
whoever
joined
just
check
the
quiz.
Okay,
you
will
be
prompting
something
like
this
and
we
are
showing
the
top
five
leaderboard.
Okay
for
the
top
five
we
are,
we
will
be
having
some
good
I
can
say
some
goodies
or
something
so
that
can
be
there
okay.
So
if
you
are
answering
the
question,
you
can
just
find
your
name
on
the
top
five.
So,
let's
start
and
all
the
best
and
yeah,
this
is
open
for
only
five
minutes,
not
more
than
that.
So
just
try
to
answer
quickly.
B
B
A
B
B
B
Yeah,
let's
see
because
there
will
be
lots
of
people
who
will
be
wearing
100,
but
let's
see
who
will
be
in
the
top
five
later
board,
because
if
you
are
answering
all
the
questions,
but
if
you
are
taking
more
time
so
it
will
already
take
you
at
the
back
okay.
So
it's
not
like
if
you're
getting
100
and
you
are
the
winner.
So
don't
suppose
that
because
I
am
checking
80
70
but
who
will
be
answering
question
at
the
starting
like
as
soon
as
possible,
who
will
be
getting
the
chance
to
win?
F
B
And
whoever
left
just
join
us,
who
is
at
the
next
time?
Okay,
we
are,
we
will
be
having
you
and
we
will
share
the
top
five
winner
for
this.
Please
on
our
social
handles
or
you
will
get
I
can
say
solution
this
by
the
end
of
today's
meeting
after
5
PM,
okay,
so
stay
tuned
for
all
the
quiz
winners
and
we
will
be
looking
out
to
you
thanks
a
lot
Michael
for
being
a
part
of
us
and
thanks
for
joining
session.
B
C
Yeah,
thank
you
Aditya.
Thank
you,
Michael
for
the
lovely
session,
so
the
next
is.
We
have
so
location.
Intelligence
plays
a
key
component
in
our
lives,
as
we
know
it
today
and
map
my
India
with
its
rich
and
extensive
25
years
of
worth.
Experience
in
this
domain
we'll
be
presenting
how
location
intelligence
can
be
comprehended
and
Incorporated
easily
into
many
real
world
problems
which
businesses
typically
face
in
today's
world.
C
So
let
me
invite
Kunal
bharti
on
the
stage
and
Kunal
bharti
is
an
associate
vice
president
at
map,
my
India,
where
he
oversees
the
development
and
management
of
the
API
products
for
iOS,
Android
web
or
any
other
form
of
SAS
software.
While
cultivating
a
market
for
the
users,
he
graduated
from
Ranchi
University
with
the
Bachelor
of
Technology
degree
degree.
He
also
did
his
Masters
in
Business
Administration
marketing
from
The
Faculty
of
management
studies
from
University
of
Delhi
yeah
hi
Kunal.
How
are
you.
G
G
G
Large
units
like
civil
aviation
operates
in
the
world
is
powered
by
an
efficient
use
of
location,
intelligence
within
their
apps
and
solutions.
So
today,
basically,
the
session
will
primarily
be
a
decision
that
I'll
be
presenting
will
primarily
be
involving.
The
main
agenda
is
how
location
what
location
intelligence
primarily
means
within
your
app,
how
you
can
leverage
it
within
your
applications
and
Tech
and
finally,
a
small
presentation
on
how
map?
G
What
is
the
role
that
mapindia
has
played
in
this
particular
field
and
easing
the
source
prop,
basically
providing
solutions
to
the
complicated
problems,
simply
simple
or
complicated
problems
which
location
brings
to
any
particular
application
for
technology,
so
without
further
Ado.
Let
me
just
quickly
continue
hello.
First
of
all
welcome
everyone,
and
thank
you
for
joining
me
and
the
team
in
this
particular
presentation.
I'll
quickly
go
through
some
of
the
important
questions
which
our
consumers
face,
which
I
have
had
the
experience
of
working
with
many
different
corporate
houses.
G
Many
different
government
organizations,
the
most
important
concerns
considering
the
recent
Dreadful
pandemic
that
we
have
been
going
through,
is
that
most
businesses
Chief
concern
is
Which
business
locations
to
reopen.
Now
this
was,
and
still
is,
a
point
of
concern
to
many
businesses,
and
this
question
is
typically
answered
by
trying
to
determine
okay,
which
places
are
safer,
which
places
are
still
under
lockdown,
which
places
are
getting
a
high
concentration
of
covert
cases
still
active
and
which
places
are
in
general.
G
Another
important
question
is
that,
okay,
this
is
quite
common
and
that
which
customer's
address
or
delivery
point
does
your
solution
cover
the
a
very
important
question
that
this
particular
very
important
problem
is
that
this
particular
solution
solves
is
okay,
I
have
let's
see,
I,
have
a
a
I,
have
a
delivery
service
I'm
a
local
dabbawala
I'm,
a
local,
basically
dabawala,
supplying
to
my
location,
I,
have
a
lunch
provider
lunch
basically
lunch
lunch
box
provider
to
my
local
community,
even
that
particular
that
hyper
local
solution
also
requires
you
to
understand.
Okay.
G
What
is
my
coverage
area?
Do
I
go
beyond
my
coverage
area?
Will
that
be
economically
feasible?
Will
that
give
me
that
return
of
investment
that
I
make
to
deliver
to
that
particular
area
or
not?
So
these
are
some
of
the
solution.
Basically,
questions
that
location,
intelligence
answers
in
your
apps
again
from
what
which
employees
are
coming
from,
where
which
area
and
what
is
your
employee
demographics
in
the
current
work
from
home
scenario,
a
very
important
question
that
the
recent
pandemic
has
forced
the
organizations
to
answer
now.
G
This
particular
question
is:
will
be
of
most
interest
to
most
of
the
people
here
that
people
get
a
choice
of
whether
they
want
to
integrate
location-based
cloud
services
into
their
apps
or
not,
but
the
decision
comes
out
comes
to
whether
it
will
provide
a
value
and
I
can
definitely
say
yes
to
most
of
the
use
cases.
Location
based
cloud
services
definitely
provide
some
value
to
any
application,
and
this
is
a
popular
misconception
that
doesn't
having
location
based.
G
Basically,
cloud
services
within
your
application
does
not,
or
is
not
not,
that
much
of
a
value
I'll
just
simply
show
from
my
presentation
that
how
you
can
provide
a
value
and
basically
just
to
justify-
and
this
this
question,
to
justify
the
solution.
Basically,
answer
to
this
question
at
how
location-based
services
do
provide
value
again,
do
you
want
your
users
to
have
a
location
enabled
nudge?
Now,
if
you
consider
several
organizations,
for
example,
several
ride
railing
Services,
they
depend
entirely
on
location-based
Services.
They
are
largely
basically
entirely
operational
through
automated
means.
G
There
is
hardly
any
staff.
There
is
only
a
small
amount
of
dedicated
calls
and
to
support
available,
but
the
entire
technology
stack
of
most
location-based
ride
healing
Services
is
powered
entirely
on
a
location-based
track.
Now,
coming
from
this,
this
deep
integration
of
location-based
services
within
their
solutions
to
a
small
use
case
where
you
simply
want
to
know.
G
I
give
an
example
of
real
life.
Example:
live
applications
in
the
real
world
now
a
store,
locator
services
is
very
popular
and
almost
all
ATMs
any
Outlets
do
provide
these
Services
consider
if
they
have
a
presence
on
the
internet
they
want
to.
For
example,
if
you
want
to
search
okay.
What
is
the
nearest
outlet
of
this
particular
this
particular
business
in
an
area?
G
Now
this
is
called
as
a
nearby
API,
which
I'll
I'll
give
you
an
example
that
mapminder
does
provide,
and
this
simple
API
or
a
simple
solution
gives
an
answer
to
how
you
can
discover
that
particular
businesses
Outlets
within
within
a
location
by
giving
the
reference
of
the
user's
current
location
a
simple
example
of
this
being
that
in,
for
example,
if
you
have
you,
if
you
want
to
search
for
ATMs
of
a
branch,
you
can
simply
search
around
the
user's
current
location.
But
these
are
the
ATMs
near
my
Branch.
G
You
can
get
a
direction
to
that
particular
ATM
by
using
what
is
called
a
routing
API,
which
gives
nothing,
but
a
directional
service,
step-by-step
Direction
service
of
Road
distance,
based
navigation
steps
from
your
current
location
to
the
destination,
the
destination
being,
whatever
nearest
ATM
that
you
have
selected.
This
is
a
live
example
of
how
store
locators
are
used
in
many
different
industry.
Use
cases
where
people
where
businesses
or
any
firm,
has
to
provide
a
discovery
ability
of
their
Outlets.
It
can
be
any
sort
of
an
outlet.
It
can
be
a
service
outlet.
G
It
can
be
a
showroom.
It
can
be
anything
to
their
consumers
fmcg
now
fast
moving
consumer
goods
typically
have
a
problem
of
providing
discoverability
on
the
basis
of
okay.
Where
do
I
deliver
my
goods
to
now
this?
There
are
a
lot
of
basically
e-commerce
platforms
which
provide
smcg
based
deliveries,
basically
Pantry
Services
Services,
which
tend
to
provide
you
deliveries
within
a
couple
of
hours
because
they
promise
the
freshness
of
the
goods
now
for
such
Services.
Typically,
the
most
common
answer,
the
most
common
problem,
that
the
face
is
the
delivery
location.
G
Now
delivery
locations
can
be
very
confusing.
Considering
that
India
does
not
have
a
very
standardized
address
system,
you
get
from
a
very
structured
addresses
in
planned
cities
such
as
gandhinagar
or
Chandigarh
to
very
unstructured
addresses
in
basically
an
urban
conglomerations
like
delhi,
now
Delhi
itself,
for
example,
Delhi
or
Mumbai
has
certain
sections
which
are
plant
sections,
which
will
have
a
very
defined
address
hierarchy
to
also
they
will
have
those
sections
of
the
city
which
will
which
won't
have
that
that
much
of
a
plant
infrastructure
or
a
plant
address
hierarchy.
Now.
G
This
is
quite
confusing
if
you
are
to
go
and
deliver
on
the
basis
of
just
a
poster
address
without
any
digital
component
attached
to
it,
without
what
is
generally
termed,
as
geocoding
now
geocoding
as
a
service.
What
it
does
is
basically
converts
a
real-life
human
readable
address
to
a
coordinate
to
a
pin
on
the
map.
G
Putting
this
pin
on
the
map
necessarily
means
simply
by
okay,
that
this
is
the
location
that
you
want
to
deliver
your
goods
to
now.
This
is
a
very
simple
example
of
how
maps
and
location-based
Tech
can
be
included
in
a
in
a
Goods
in
an
e-retail
business,
fmcg-based
e-detail
business,
wherein
you
can
get
an
assurance
that,
since
you
have
the
exact
position
on
the
map,
so
you
can
definitely
go
to
that
particular
location.
G
One
by
having
from
the
pin
on
the
map
the
other
way
around
by
deriving
the
address
from
the
pin
on
the
map
or
by
entering
the
complete
address
and
getting
the
lat
and
getting
the
latitude
longitude
or
the
geocode
geocordinate
distribution
to
fetch
the
address
from
the
pin
on
the
map.
The
the
mechanism
is
reverse
decoding,
which
is
the
other
way
around,
simply
by
taking
the
coordinates
as
an
input
and
providing
a
full
human
readable
address,
and
the
geocoding
as
I
have
already
explained
this.
G
Another
example
is
that
people
tend
to
provide
a
field
wherein
they
can
basically
type
any
address
and
the
search
results
come
in
a
list
and,
as
you
are
typing,
the
search
results
get
basically
gets
refreshed
on
the
more
characters
that
you
type
in.
They
get
more
filtered
out,
and
you
get
to
your
you
get
the
point
that
you
get
your
desired
address
in
the
list
and
you
simply
select
from
the
list
and
your
address
is
detected.
G
No.
This
is
what
is
called
typically
called
an
autocomplete
or
an
auto
suggest.
Api
again,
all
of
this
all
of
these
Services
fmcg
fmcg,
based
apps,
typically
have
multi-platform
support.
They
do
provide
Maps
based,
basically
pin
marker
based
support
on
their
web
applications,
their
Android
or
iOS
applications
and
location-based
services
are
typically
available
for
all
of
these
platforms.
G
G
Now
again,
an
example.
A
real
life
example
of
food
delivery
and
on-the-go
services
are
also
one
of
the
use
cases
in
the
industry
where
location-based
Services
have
started
playing
a
key
component.
Key
role
is
that
when
you
order
a
food
from
any
on-the-go
Services,
you
tend
to
get
two
aspects.
You
first
need
to
provide
your
address.
G
It's
again
goes
to
the
same
mechanism
that
you
either
provide
and
address
to
geocoding
or
pin
on
the
map,
and
the
second
component
is
that
you
are
able
to
see
in
a
map
the
food
coming
to
your
location.
You
are
able
to
see
basically
what
is
called
a
live
track
of
the
delivery
boy
or
The
Delivery
Agent,
coming
to
your
location.
Now.
This
is
a
very
good
example
of
providing
real-time
feedback
through
the
use
of
location-based
services
to
your
consumers.
G
Why?
Because
this
provides
an
ease
of
Mind,
a
very
good
user
experience
to
your
consumers,
that,
yes,
there
is
a
transparency
that
I
can
see
in
real
time
that
the
Delivery
Agent
is
indeed
coming
to
my
location
and
if
he
or
she
Strays
from
his
or
her
location,
I
can
definitely
call
call
him
or
her
up
and
do
a
course
correction.
G
G
They
tend
to
provide
a
specialized
specialist
subset
of
nearby
search,
which
is
called
a
point
of
Interest
along
the
Route
search,
but
rather
than
searching
along
around
a
particular
reference
points,
searches
along
the
route,
for
example,
if
your
Delivery
Agent
is
already
following
a
route
from
point
A
to
point
B,
and
he
needs
to
pick
up
basically
from
the
restaurant,
which
falls
on
his
route.
He
can
simply,
basically
this
the
application
at
the
back
end
can
simply
search
okay.
G
This
is
the
restaurant
that
falls
along
your
route,
which
has
the
ability
to
fulfill
the
said
order
of
the
user,
and
he
can
simply
select
that
particular
the
system
can
simply
select
that
particular
Outlet.
The
driver,
then,
can
simply
pick
up
from
that
particular
outlet
and
proceed
towards
the
destination.
G
This
is
the
point
that,
with
the
store
is
particularly
operational
at
and
you
can
basically
go
to
that
particular
store,
and
if
it
is
Falls
outside
of
the
time
window,
then
you
can.
You
may
not
time.
Windows
are
one
of
those
things
that
are
provided
through
the
place
detail
of
apis,
but
it's
not
one
of
the
only
things
there
are
a
lot.
Many
details
depends
upon
use
case
to
use
case.
Restaurants
typically
have
time
Windows
they
have
what
type
of
which
basically
Foods
or
services
that
are
available
through
them.
G
For
example,
people
in
the
current
use
case.
Do
they
follow
the
latest
safety
standards?
Do
they
do
they
have
vegetarian
food?
Do
they
have
non-vegetarian
food?
Do
they
have
if,
for
example,
if
they
have
a
menu-based
Services
okay,
do
they
provide
Continental?
Do
they
provide
Chinese
cuisine
types?
So,
basically,
all
of
these
services
are
typically
related
to
Place
detail
API
when
it
comes
to
location-based,
Tech
and
again.
Finally,
if
you
have
already
want
to
know,
this
is
basically
to
know
which
is
the
nearest
outlet
that
can
serve
a
particular
user's
location.
G
Another
example
of
the
same
situation-
basically
this
particular
in
this
particular
use
case.
As
you
can
see,
your
your.
The
application
is
simply
asking
the
user
to
confirm
his
or
our
location
on
the
basis
of
his
confirmed
location.
His
deliver
delivery
store
will
be
confirmed
now.
E-Commerce
again,
e-commerce
as
a
service
also
has
many.
G
This
particular
use
case
takes
care
of
many
different
types
of
apis
from
any
basically
location-based
apis
as
such,
of
course,
Maps
is
primarily
one
of
them,
reverse
geocoding,
again
quite
important,
and
it
may
interest
you
that
any
UPI
based
transactions
in
India
they,
the
UPI
applications,
are
actually
bound
to
capture
where
the
UPI
transaction
took
place.
So
they,
whenever
you
transact,
whenever
you
do
a
transaction
on
UPI,
the
applications
tend
to
capture
your
location,
information,
they
capture
the
latitude
and
longitude.
G
This
is
mandated
by
a
government,
basically
policies
and
since
they
capture
the
location
information,
they
are
also
able
to
tell
basically,
through
the
use
of
reverse,
recording
API
capturing
the
latitude
longitude
converting
into
the
address
that,
where
actually
the
transaction
took
place.
So
e-commerce
as
a
platform,
basically
captures
your
location
so
that
they
can
tell
you,
which
are
the
retail
outlets
which
are
applicable
to
you,
which
of
the
retail
outlets,
can
actually
serve
to
you.
G
If
you,
if
you
notice
that
money
e-commerce
applications
tend
to
ask
you
for
your
PIN
code
and
on
the
basis
of
the
PIN
code,
they
are
able
to
tell
that
which,
basically,
which
of
the
offerings
on
their
platform,
is
deliverable
to
your
location
or
not.
This
is
through
the
use
of
a
deeply
integrated
location-based
Tech
at
their
backend
to
the
use
of
which
they
derive.
This
particular
information
that
whether
a
particular
good
or
a
particular
type
of
a
service
is
available
to
the
consumer
or
not.
G
Again
an
example:
they
do
ask
for
your
delivery
address,
so
that
is
of
course,
provided
by
either
a
pin,
learner,
location
or
a
geocoding
space
search
after
you
have
basically
ordered
a
particular
location.
Some
e-commerce
applications
also
allow
you
to
live
track,
your
shipment,
basically
to
see
where
it
has
reached,
and
so
that
you
can
basically
see
under
the
off
seat
on
the
map
that
okay,
it
has
been
shipped
it
is.
It
has
been
shipped
from
your
Warehouse.
G
G
Payment
again,
as
I
said,
UPI
already
has
a
very
important
policy
that
it
does
capture
your
location
information.
Apart
from
that,
you
may
also
notice
the
certain
payment
apps
provide
you.
What
are
the
nearest
Merchants
which
support
their
services,
so
this
particular
feature
is
also
taken
care
of
by
location
based
Tech
by
location-based
apis
such
as
reversio
code
and
maps.
G
Of
course,
as
you
can
see,
one
of
one
of
the
most
popular
one
of
the
largest
payments
app
in
India
does
use
map
menu
based
apis
map
apis
and
sdks,
wherein
they
can
provide
you.
What
are
the
nearest
businesses,
depending
upon
the
different
categories
that
they
have
and
which
business
categories
of
among
them
actually
take
that
particular
payment
platform
as
one
of
the
payment
services.
G
Few
more
screens,
which
show
you
the
same
now
Automotive
is
another
sector
where
location
based
Tech
has
long
been
involved,
and
not
only
to
provide
basically
navigation
from
one
place
to
another,
and
this
was
the
Legacy
mechanisms
that
typically,
when
we
talk
about
Automotive
sector,
we
used
to
think
about
okay,
location
based
Tech
in
an
automatic
automotive
sector.
Typically
only
means
means
providing
navigation
services
from
one
point
to
another.
G
Now
Automotive
has
expanded
and
almost
all
the
vehicles
that
you
see
right
today
are
more
connected
to
the
internet
than
they
were
now
since
they
are
connected.
They
are
always
connected
to
the
internet.
They
can
provide
a
lot
more
value
addition
to
the
user,
for
example.
They
can
provide
you,
okay.
How
much
time
did
you
spend
on
the
road
moving?
How
much
time
did
you
spend
while
idly?
How
much
time
did
you
spend
while
the
ignition
was
on
and
you
were
standing
at
that
stop
line?
G
Maybe
you
could
have
switched
your
ignition
off
and
saved
X
percentage
of
fuel,
so
all
of
these
Services,
which
take
into
account
that,
where
all
you
have
been
provide,
a
real-time
driver
analytics
provide
tracking
Based
Services
in
case
of
a
stolen
vehicle.
They
can
also
provide
services
like,
for
example,
okay.
You
have
parked
at
this
location,
where
have
you
parked
your
vehicle
so
that
you
can
easily
find
it,
especially
if
it's
a
large
parking
location?
G
All
of
these
Services
typically
use
a
location-based
Tech
in
the
back
end
and
all
of
you,
as
you
can
see,
they
provide
a
very
intuitive
and
a
good
user
experience
through
the
use
of
location-based
Tech.
They
provide
an
extreme
value
addition,
and
some
of
some
of
these
components
are
indeed
marketed
by
automotive
apps,
as
their
unique
selling
points
to
give
them
a
competitive
advantage
over
their
competing
grants.
G
Again,
one
of
the
good
examples
that
we
recently
have
is
due
to
the
pandemic,
one
of
the
largest
luxury
car
manufacturers
in
India.
It
also
offered
basically
Co
basically
covet
based
information
within
their
own
infotainment
systems.
Now,
infotainment
system
is
the
India
system
that
you
typically
get
in
a
car
where
you
see
navigation,
other
such
components
already
being
implemented
there.
Now,
apart
from
that,
they
also
offer
things
like,
which
is
the
nearest
Kobe
Treatment
Center,
which
is
the
nearest
kobit
test
center
and
such
enhanced
services.
G
Such
extensibility
of
infotainment
system
is
only
possible
if
you
have
a
very
deep
and
a
very
excalibal,
basically
connection
to
any
location
based
cloud
services
provider,
for
example,
this
particular
luxury
car
manufacturer
has
has
made
had
did
a
partnership
with
mapind
you
to
provide
this,
these
services
on
the
go,
and
so
even
in
their
already
sold
Vehicles.
This
was
simply
a
software
update
which
they
did
in
the
background
and
all
the
all
the
vehicle
owners
started
to
get
this
information
streamed
onto
their
headsets
or
their
infotainment
systems.
G
Another
example
which
I
already
told
is
the
ride.
Hailing
Services.
They
also
use
almost
all
of
location
based
decks.
This
is
a
very
deeply
integrated
use
case,
wherein
almost
all
ride.
Hailing
Services
you
search
through
the
use
of
Auto
syst
API.
They
use
maps,
they
use
maps
sdks
to
display
maps.
They
use
routing
functionality
to
give
driving
directions
to
the
user.
They
have
inbuilt
navigation
to
provide
to
so
that
their
drivers
can
drive
from
point
A
to
point
B
and
all
of
these
Services.
G
A
good
example,
if
you
must
have
noticed
Cohen
website,
also
had
a
very
deep
integration
with
location
based
Tech
how
they
used
to
provide
discoverability
of
basically
vaccination
centers
was
through
the
help
of
location-based
components.
They
either
provided
you
to
search
your
vaccination
centers
by
pin
code.
They
either
provided
to
you
to
search
your
vaccination
centers
by
district
and
also
to
search
your
vaccination
centers
by
map
in
which
map
India
had
a
contribution
towards
the
coven
government
initiative.
G
So
all
of
these
companies,
as
you
see
even
the
government
for
small
small
things
for
for
such
important
things,
small
small.
These
are
just
small
small
components
which
in
turn,
make
a
very
large
impact
to
the
consumer,
and
this
was
a
very
good,
basically
initiative
done
by
the
government
to
provide
such
such
location-based
search
Services,
because
it
makes
a
lot
more
sense
for
people
to
discover
if
they
already
know.
Okay,
what
is
the
district?
G
What
is
the
PIN
code
and
if
they
don't
know
their
pin
code,
that
they
can
simply
search
the
location
on
the
map
or
on
the
basis
of
the
user's
current
location,
they
can
simply
search
what
are
the
nearest
vaccination
centers
and
what
is
the
status
of
the
said?
Vaccination
centers?
No,
as
you
can
see,
in
almost
all
the
use
cases
that
we
have
presented
till
now,
they
have
a
very
deep
connection
to
how
location
based
Tech
is
used
now
continuing
on
skip.
G
G
Now
all
of
this,
the
one
thing
that
can
tie
all
of
these
different
components.
These
intermediaries,
the
folks
and
the
large
firms
together,
we
think,
is
apps
apps
as
a
component
and
the
latest.
If
latest
GIS
and
tools
developed
on
Rich
maps
are
utilized
within
your
apps
to
make
them
location
aware
they
provide
a
great
value.
They
provide
a
lot
of
deep
insight
into
how
your
applications
have
performed
in
which
locations
they
provide
analytics
to
you,
as
well
as
enhance
the
user
experience
for
your
users.
G
People
tend
to
associate
that
mapping
is
quite
complex.
Yes,
mapping
is
complex,
but
using
mapping.
Solutions
is
quite
easy,
as
you
have
already
seen
from
the
way.
Almost
all
of
these
use
cases
have
deeply
integrated
location.
Technologies
using
mapping
is
easy.
It's
just
that
you
need
to
understand,
okay,
which
component
of
mapping
is
required
where,
and
once
this
Clarity
is
with
you,
then
simply
picking
up
and
choosing
from
the
wide
variety
of
mapping.
Solutions
is
a
piece
of
cake,
and
this
is
where
map
India
as
a
platform.
G
Actually,
as
a
firm
can
help
you
mapping
your
maps
and
services
are
on
everything
based
location-based
Tech
are
available.
We
have
easy
to
integrate
tools
to
empower
any
location-based
modules
for
applications
and
website.
I'll,
just
take
a
couple
of
minutes
time
to
quickly
scroll
through
the
different
services
that
we
have
available.
G
We
have
a
rich
API
Library,
which
can
basically
you
can
use
to
do
an
interactive
map
display
you
can
do
accurate
building
level
geocoding,
you
can
do
reverse
geocoding.
You
can
provide
turn
button
directions
and
guidance
to
your
to
your
consumers.
You
can
provide
nearby
POI
search.
You
can
do
basically
type
as
you
go.
Auto
suggest,
search,
integrate,
Auto,
says,
search
apis
in
your
applications.
G
Almost
all
of
our
Rich
map
data
is
all
of
it
is
hosted
on
the
AWS
platform
and
thanks
to
the
AWS
team,
it's
highly
scalable.
It
is
a
highly
secure
and
we
have
multiple
platforms.
We
have
our
mapping
platform,
we
have
our
iot
platform.
We
also
have
recently
come
up
with
our
own
analytics
platform,
which
all
of
these
power
are
sdks
and
apis,
which
in
turn,
power
our
own
products
as
well
as
third
party
applications.
G
Some
of
the
map
sdks,
which
are
available
from
us,
are
in
basically
the
map
sdks
for
your
web
map.
Sdks
for
your
fluidic
map
integration
in
your
mobile
map,
mobile
apps,
we
have
navigation
SDK,
which
is
a
unique
selling
point
for
map
Mania,
that
we
are
the
one
of
the
sole
providers
of
navigation
SDK,
which
provides
navigation
embedded
within
your
app
rather
than
calling
out
a
third-party
app.
Just
give
me
one
moment.
G
G
Okay,
where
does
the
map
is
map?
Sdks,
usually
help?
They
provide
visual
awareness
to
the
user's
location
they
allow
for
animated
visualizations
about
your
services
on
the
map
and
they
help
correlate
virtual
references
to
the
physical
twin
in
the
real
world.
This
is
how
Maps,
as
a
map
display
as
platform
as
a
service
basically
provides
value
to
any
use
case.
G
Now
we
have
our
search
apis,
which
basically
Turn
The
Real
World
locations
easily
and
basically
to,
and
it
provides
varying
filters
to
narrow
down
the
search
and
put
a
place
on
the
pin
on
the
map.
We
have
the
type
as
you
go
search.
We
have
the
geocoding
API.
We
have
reverse
opening
API
the
nearby
API,
which
searches
for
nearby
places
according
to
a
given
category
in
a
location
we
have
Place
detail
apis.
G
We
have
text
search
apis,
which
tend
to
take
in
a
lot
of
which
tend
to
incorporate
a
lot
of
machine
learning
to
make
a
show
that
even
vague
statements
can
be
converted
into
geographical
components,
and
we
have
what
is
called
a
network
standardization
EPA,
because
India
has
a
address.
Structure
is
quite
half
Hazard,
so
we
have
an
interest
rendition
API,
which
can
transform
postal
address,
even
if
it
is
non-standardized
into
a
standardized
admin
hierarchy.
G
Where's
this
health,
basically
search
Services,
typically
help
because
users
tend
to
know
their
location
in
the
form
of
postal
addresses,
but
to
resolve
them
into
digital.
Addressing
is
where
the
search
apis
actually
worked
in
magic,
routing,
apis
map
menu,
routing
apis,
basically
solve
the
complex
routing
problems
for
fleets,
Enterprises
and
individuals.
They
provide
comprehensive
and
updated
accurate,
turn-by-turn
navigation
between
two
or
more
geolocations,
some
of
the
roots
and
navigation
apis
that
we
have
is
rooting
apis
for
passenger
vehicles
for
heavy
Vehicles.
We
have
our
own
navigation
SDK.
G
As
I
mentioned,
you
can
embed
navigation
within
your
app
rather
than
calling
out
another
application
for
navigation.
We
have
the
distance
Matrix
API,
which
provides
driving
distance
and
duration
between
two
or
more
points
with
or
without
live
traffic
consideration.
We
have
routing
apis
for
two
wheelers
again.
All
of
these
apis
can
take
into
account
live
traffic
information
which,
basically
is
provides
congestion
addition
to
your
etas
and
basically
provides
real-time
ETS
on
the
basis
of
the
actual
congestion
of
the
route.
G
We
have
snap
to
Road
apis,
which
basically
map
matches
any
raw
coordinates
for.
This
is
quite
useful
in
telematics
use
cases
wherein
you
get
GPS
coordinate,
which
is
not
matched
to
the
map
road
Network.
So
this
is
actually
matches.
The
raw
GPS
coordinates
to
the
road
networks,
the
travel
root
image
API
and
the
PO
along
the
API,
which
basically
searches
for
here
along
the
Route.
G
Now
how
this
helps
is,
basically,
it
helps
gives
driving
direction
to
any
place
to
your
users.
It
calculates
the
driving
time
or
distance
based
reports
to
planned
logistical
operations
in
your
Solutions.
You
can
also
do
driving
range
based
polygons
on
the
basis
of
expected
distance
or
travel
time.
You
can
map
match
your
raw
GPS
location
to
Road.
Network
I'll
quickly
skip
some
part
of
it.
G
The
genetics
apis
basically
provides
analytical
solutions
to
geographical
queries,
basically
visualization
on
the
basis
of
certain
geographical
queries,
for
example,
draw
all
the
districts
in
let's
say,
uttarakhand,
where
the
population
exceeds
a
certain
parameter.
So,
on
the
basis
of
certain
demographic
or
socioeconomical
values,
you
can
do
basically
analytical
operations
and
draw
the
said
geographical
admins
on
the
basis
of
input
queries.
G
Now,
our
search
apis
are
basically
a
bit
customized
to
the
recent
situations
on
the
basis
of
all
the
recent
keywords
that
are
more
more
basically
contribute
more
towards
the
recent
what
you
say
happening
in
the
recent
situations
around
the
world.
We
tend
to
make
our
keywords
more
Dynamic,
so
our
search
is
more
real
time,
for
example.
G
Well,
if
you
simply
type
in
covet
since
covet
is
a
very
ongoing
problem
at
this
moment,
so
we
tend
to
provide
more
search,
results
related
to
covet
and
simply
some
some
simply
some
some
types
of
keyword
code.
So
our
search
is
more
use,
more
keyword,
driven
on
the
basis
of
the
more
popular
keywords
according
to
the
current
situation
and
so
and
so
forth.
B
Yeah
I
yeah,
so
I
think
yeah.
So
we
are
having
lots
of
questions.
So,
okay,
let
me
introduce
our
something.
Is
something
important
is
left
because
there
are
so
much
good
questions
out
there.
So
let
me
ask
it
one
by
one
so
like
yeah,
thanks
for
the
great
explanation
and
I
can
see
a
lot
of
good
comments
in
the
comment
section
and
those
are
so
awesome.
Okay,
so
people
have
questions
like
all
like
all
our
DCPS
are
free
for
use.
G
This
is
the
freemium
service.
Some
of
the
apis
are
free
for
use
up
to
a
particular
amount
of
transactions,
and
after
this,
basically
you
get,
you
can
basically
opt
for
more
transactions
on
the
basis
of
a
business
agreement
with
mapindia.
It
typically
depends
upon
your
use
case
to
use
case,
so
our
Solutions
is
tailored
according
to
use
cases.
If
you
have
a
particular
use
case,
you
can
contact
us
at
our
website.
It's
mapinder.com
API,
you
can
contact
us
and
if
you
have
any
basically
questions
related
to
commercials.
B
G
Okay,
I
can
provide
you
the
most
recent
ones.
So
coven
is
one
such
use
case
where
government
definitely
uses
our
use
our
apis,
and
if
you
have
perhaps
heard
of
umang
now,
umang
is
a
digital
application
platform
by
the
government
of
India.
It's
powered
by
the
digital,
India,
Foundation
I
think,
and
it
has
a
lot
of
services
such
as
it
has.
H
G
That
there
are
many
other
basically
services.
For
example,
dgca
uses
our
services
for
drone
registration
systems
in
India,
which
is
director
general
of
civil
aviation
uses
Services,
provided
by
map
India,
the
gstn,
which
is
basically
goods
and
services.
Tax
net
tax
system
also
uses
our
services,
and
many
such
government
departments
such
as
Telangana,
police
and
I,
can
basically
go
on
saying
the
names
and
so
and
so
forth,
but.
B
Many
many
different
government
organizations-
yeah
yeah
I,
got
it
thanks
a
lot
for
this
explanation
and
the
next
question
we
are
having
like
what
if
these
Services
is
provided
in
ambulances,
where
the
person
and
the
driver
can
track
each
other
location.
So
this
is
something
like
idea,
so
if
this
can
be
implemented
with
using
the
web
and
apis,
yes,.
B
G
G
Since
we
are
lack
of
time,
then
you
can
go
to
visit
our
website,
mapping.com
API
and
the
questions
that
we
have
are
such
for,
like
you
have,
for
example,
if
you
have
any
use
case
now,
there
are
a
couple
of
questions,
the
first
one
being
that
in
all
the
use
cases
that
we
presented,
how
do
you
think
that
location
based
Tech
can
be
leverage
more
efficiently?
How
can
we
leverage
more
efficiently
and
how
it
can
be?
G
How
more
of
the
more
of
the
location-based
Services
can
be
implemented
in
in
a
particular
use
case,
and
how
do
you
deem
that,
according
to
your
understanding
that
location-based
Services
provide
value
to
that
particular
use
case?
That
was
question
number
one,
and
this
question
number
two
is
in
a
particular
use
case
of
your
choice.
You
can
pick
use
case
according
to
your
industry,
according
to
your
particular,
basically
any
application
that
you
have
had
to
work
on
or
any
application
or
a
solution
that
you
have
had
a
chance
to
explode.
G
This
is
a
particular
use
case,
driven
question
that
if
you
have
let's
say,
if
you
have
a
store,
locator
Based
Services
store
locator
based
service,
wherein
that
you
have
to
provide
you
have
to
integrate
location
based
Tech
from
map
India,
which
of
the
apis
and
sdks.
Would
you
use-
and
you
have
to
make
sure
that
this
is
a
store
locator
based
service,
wherein
you
have
to
also
provide
Field
Force
analytics
component
as
well
as
in
that,
whenever
your
agent
goes
to
that
particular
store?
G
You
have
to
basically
monitor
that
particular
user
and
give
an
idea
that
okay,
how
do
I,
how
would
I
monitor
that
particular
user?
How
would
I
get
so,
and
so
information
of
that
particular
user?
So
here
you
have
to
tell
us
what
all
information
you
can
derive
from
which
of
the
apis
and
sdks
provided
by
map
India,
and
for
that
you
can
visit
our
website,
mapping.com
API,
to
know
more
about
the
apis
and
services
provided
on
map
India.
G
B
Yeah,
that's
a
great
expression
and
let
me
sum
it
up
for
you
all:
the
participants
are
there.
So
what
we
have
done
is
we
have
just
checked
the
meter
description.
There
is
a
Google
form
is
already
there,
okay,
so
what
you
have
to
do
just
so,
as
the
three
questions
already
explained,
so
what
you
have
to
do
just
think
of
an
idea
where
you
can
use
a
webinar
apis
or
if
you
have
a
any
of
the
startup
idea
or
where
you
can
use.
Let's
say
you
are
like
Devastation
for
ambulance.
B
So
if
you
are
assuming
that
ambulance,
if
you
will
use
mapa
ndib,
so
how
that
tend
to
help
you
to
all
the
people?
Okay,
so
you
can
think
of
these
ideas
and
just
sum
up
the
idea
with
a
great
conclusion.
So
description
should
be
a
lot
of
words.
Okay,
so
that
vertical
description
you
have
to
write
in
I
can
say
in
comment
section:
there's
a
Google
form
already
there,
so
you
can
find
it.
My
India
and
I
can
see
surprise
space
name,
okay,
so
the
Google
Formula
is
there.
B
The
Google
form
is
open
till
8
pm
today.
So
you
have
to
fill
it
out
with
all
your
response
and
tomorrow
you
will
be
having
a
session
again
with
map.
My
India
for
the
I
can
see
10
to
20
minutes.
Then
they
will
be
announcing
all
the
winners
for,
like
all
the
what
I
can
say,
what
are
the
questions
we
are
getting
so
we'll
be
running
winner
tomorrow?
Okay,
so
hope
it
answers
your
question.
B
D
C
C
I
Sure
so,
like
Seema
introduced
me
I
work
and
they're
prone
as
a
community
solution.
Architect,
I
have
around
10
years
of
experience.
So
in
the
industry
at
devtron
lab,
we
have
created
a
end-to-end
software
delivery
workflow
for
kubernetes,
and
it
is
made
on
kubernetes
itself,
so
I'll
be
sharing.
My
screen
is
the
screen
visible.
B
I
I
That
is
why
we
created
the
open
source,
different
solution
which
integrates
with
most
of
the
major
solutions
that
are
present
in
the
kubernetes
markets
Marketplace
and
while
giving
you
all
the
flexibility,
we
also
give
you
multiple
abstractions
on
on
the
level
that
you
won't
need
to
go
into
the
specifics
of
kubernetes.
While
we
still,
you
will
be
able
to
deploy
applications
and
productionize
it
using
their
prawn,
using
a
completed
complete
end-to-end,
automated
workflow
okay.
I
So
there
are
new,
numerous
tools
that
we
use,
that
we
integrate
with
different,
that
are,
that
can
be
integrated
with
devtron.
Some
of
them
are
rocd.
We
use
Clear,
we
use
Hampshire
charts,
so
that
was
one
of
the
primary
concern.
Why
we
created
devtron,
broken
user
management.
Is
one
such
thing
that
kubernetes
Community
is
very
skeptic
about,
so
the
user
Management
in
kubernetes
is
pretty
much
a
very
basic
and
we
take
it
to
the
next
level
by
integrating
stuff
with
Caspian
with
different.
I
So
the
solution
covers
integrates
with
multiple
services
within
the
kubernetes
and
cncf
space,
and
we
are
more
focused
on
giving
a
better
experience
to
the
users
who
are
using
devtron
and
not
and
and
the
people
should
not
be
wandering
here
and
there
to
search
and
installs
install
the
stuff
on
their
kubernetes
cluster.
So
we
do
it
at
one
place
all
at
one
place
so
right
from
monitoring
stuff.
We
give
you
a
solution
for
deployments.
We
give
you
a
solution
for
githubs.
I
We
give
you
a
solution
for
observability
of
your
applications,
so
everything
is
integrated
in
one
good
package
that
you
can
install
and
or
right
away
deploy.
Applications
on
production
environments,
I
would
be
I
would
not
be
going
into
the
details
of
for
the
architecture,
but
basically
we
have
around
14
to
15
micro
Services.
I
Some
micro
services
are
like
open
source
solution
that
are
readily
available
in
that
are
really
available
in
the
open
source,
Marketplace
and
some
solution
them.
Some
micro
services
are
created
by
their
front
itself
and
it
is
in
open,
open
source
space
itself,
so
you
can
just
go
ahead
and
poke
the
devtron
repository
on
GitHub
and
you
can
even
contribute
to
the
devtron
as
it
is
the
open
source
solution,
completely
open
source
solution,
there's
no
premium
or
ifs
and
butts.
I
I
You
can
add
any
Helm
repository
that
exists
and
all
we
import
all
the
charts
from
that
particular
hem
repository
and
we
make
it
available
for
you
to
install
and
just
you
can
use
any
health
chart
within
a
click
of
within
a
click
of
your
mouse
and
get
it
ready
and
get
it
deployed
at
any
of
your
kubernetes
clusters.
We
support
multiple
deployment
strategies,
blue
clean
and
Cannery
deployment
being
one
which
is
not
supported
out
of
the
box
in
the
kubernetes
deployments.
We
support
multiple
clusters
and
multi-cloud
support.
I
Also
so
in
one
different
installation
you
can
have
a
kubernetes
cluster
from
multiple
Cloud
providers
and
even
on-premise
integrated
within
one
different
solution.
Right
so
and
we
have
integrated
observability,
like
I
told
you,
you
don't
have
to
have
a
specific
tool
integrated
along
with
different
to
use
that
uniform
access
management
is
handled
with
another
project,
which
is
the
Caspian
project.
I
So
we
use
caspin
to
provide
you
a
unified
and
extensive
user
management
system
where
you
can
create
groups
for
your
environments
and
then
give
access
to
users
using
those
groups
which
I
will
be
demonstrating
to
you
in
a
while,
and
then
we
have
some
specific
features
like
centralized,
caching,
hibernation
and
our
her
security
policies.
That
would
be
so
I
would
be
quickly
coming
to
the
Hands-On.
I
will
be
skipping,
the
installation
part,
which
is
pretty
much
documented.
I
I
Any
containerized
application
on
a
kubernetes
cluster
so
for
the
demo
I
will
be
covering.
So
this
is
the
devtron
UI
that
we
have
I
will
quickly
cover
some
features
that
we
I
just
want
to
show
you,
so
you
can
connect
various
kubernetes
clusters
on
this.
Therefore
UI
that
you
have
so,
if
you
see
I,
have
various
clusters
connected
to
my
this
web
installation
and
then
I
can
create
several
environments
within
the
clusters.
I
So
I
can
have
multiple
non-prod
environment
running
within
the
same
cluster
and
while
still
having
a
logical
partition
of
environment
and
in
kubernetes
lingo,
we
call
it
namespace.
So
each
environment
is
mapped
to
a
namespace
within
the
same
cluster.
So
I
I
need
not
have
a
I,
do
not
have
multiple
non-prod
environment
setup,
I
just
need
to
connect
one
cluster
and
therein
I
can
have
my
Dev
environment,
my
stage
environment
and
the
pre-prod
environment.
I
I
I
That
we
have
on
this
cluster
and
then
you
can
specify
you
can
add
the
users
to
these
groups
using
the
UI
itself.
So
I
can
add
any
person
to
a
specified
group.
I
can
also
sort
of
give
a
specific
application
permissions
or
to
a
user.
I
can
also
do
that
as
well.
Moving
on
to
the
Chart
section,
which
is
the
Hem
chart
section
so,
like
I
told
you
you
can
import
any
Helm
chart
available
in
the
market
and
the
devtron
will
provide
you
that
Helm
chart
on
this
UI.
I
You
can
just
click
on
any
of
the
helm
chart
and
just
click
on
deploy.
You
can
give
it
a
name.
You
can
select
the
project
and
you
can
select
the
environment
you
want
to
deploy
it
in.
You
can
refer
the
read
B
file
from
this
side,
Pane
and
that
we
have
where
you
will
get
all
the
information
about
that
particular
Ram
chart,
and
then
you
can
quickly
deploy
it
in
any
of
your
environments.
One
more
thing
that
we
have
over
here
is
like.
I
We
also
have
a
functionality
where
you
can
group
a
certain
Health
charts
so,
like
I
have
created
this
particular
infra
Essentials
group
of
film
chart
which
I
install
in
every
cluster
that
I
attach
it
to
the
devtron.
So
this
particular
chart
group
has
the
Prometheus
Cube
from
it
is
stack
for
monitoring.
It
has
a
cluster
Auto
scaler
for
infra
Auto
scaling.
Then
it
has
a
matrix
server.
Then
it
has
a
fluent
bit
for
logging.
Then
it
has
nginx
Ingress
controller
for
all
the
ingresses
that
we
have
then.
I
So
you
can
configure
and
create
groups
like
these
and
deploy
it
in
one
click.
You
can
do
that
with
devron
itself,
so
that
is
pretty
much
the
chart
section.
Then
we
have
a
security
integrated
within
this
devtron
UI.
So
you
can
scan
all
your
images
that
you
are
building
through
devtron.
You
can
even
control
the
security
policy
at
your
organization
level.
Even
you
can
do
that
at
cluster
level
or
environment
level
or
even
add
the
application
Level.
You
can
do
that.
I
So
if
you
find
that
you
don't
want
to
you,
don't
want
any
user
to
deploy
any
a
particular
application
which
has
a
critical
vulnerability
detected.
You
can
do
that
and
then
post
that
using
these
security
policies
itself.
So
that's
that's
pretty
a
new
feature
that
we
recently
introduced.
Okay,
then
I'll
quickly
move
to
deploying
a
micro
service
on
kubernetes
classes
so
for
this
I
will
get
to
the
application
section.
So
these
are
the
micro
services
that
are
deployed
through
depth.
So
I'll
quickly
tell
you
how
to
do
that.
I
So
for
deploying
any
micro
service
using
devtron,
you
have
to
create
an
application,
so
I'll
be
giving
a
name
for
this
AWS
demo
I'll
be
selecting
a
new
project.
I'll
be
selecting
the
demo
project.
Then
you
can
either
select
a
template
means
any
app
that
you
want
to
clone
any
existing
app
that
you
have
already,
which
is
very
similar
to
your
present
application
that
you're
onboarding.
I
You
can
select
that
application
and
it
will
clone
all
your
configurations
from
that
particular
app,
but
for
this
demo
I
will
be
creating
it
from
the
scratch
itself.
So
I
have
selected
the
blank
app
okay.
So
I
have
created
this
application
so
now
I
can
select
where
my
code
is
so.
My
code
in
this
case
is
in
our
GitHub
public
repository,
which
is
this
one,
so
I'll
be
quickly
copying
this
https
so
URL
to
clone
my
repository,
so
this
I
have
put
over
here.
I
If
you
have
a
configuration
where
you
use
the
code
from
a
different
repository
and
then
you
have
configuration
set
in
a
different
gate
repository,
you
can
use
the
this
checkout
path
and
you
can
add
multiple
git
repositories
within
the
same
application
itself,
or
in
our
case
we
have
only
one
that
we
are
building
from
so
I'm,
just
adding
one.
Otherwise
you
can
add
multiple
git
repositories
over
here,
just
next
part.
Okay.
I
So
here
you
have
to
provide
Docker
repository
where
you
will
be
pushing
all
your
images
once
the
build
are
done
so
I
will
be
just
pushing
it
to
different
our
repository
that
I
have
created
for
this
one,
so
I
have
to
mention.
Where
is
my
Docker
file?
So
if
I
can
show
you
over
here,
so
I
have
my
Docker
file
in
SRC.
Slash
application-
and
this
is
the
this-
is
my
daughter
file
right.
So
I
will
be
mentioning
this
part
SRC
slash
app.
I
I
So
I'll
save
this
configuration
moving
on
to
the
next
step.
This
is
the
deployment
template.
So
this
is
the
actual
configuration
file.
This
is
the
one
configuration
file
that
devtron
uses
and
it
configures
everything
for
you.
So
if
you
want
your
application
to
Auto
scale,
this
is
the
place.
You
can
configure
that
if
you
want
for
disruption
budgets
to
be
enabled
for
your
application,
this
is
the
place.
You
can
do
that
if
you
want
some.
I
If
you
want
to
pass
some
arguments,
if
you
want
to
pass
some
environment
variables,
you
can
do
that
over
here.
So
I'll
quickly
show
you
so
this
configuration
is
sufficient
to
work
for
most
of
the
common
applications.
You
just
have
to
mention
your
port
number
over
here
and
then,
if
you
want
to
configure
some
configurations
over
here,
you
can
do
that
in
our
case,
I
won't
be
using
using
HTTP
2,
so
I'm
just
disabling
it
over
here.
I
So
I'll
quickly
walk
you
through
or
some
configurations
that
are
very
essential
and
for
other
configurations
you
can
refer
to
this
documentation.
That
is
already
there
on
top.
So
it's
it
takes
you
to
the
docs.devtron.ai
and
then
here
are
every
each
and
every
configuration
that
I
have
in
the
deployment.
Template
is
a
documented
pretty
well
over
here.
So
you
can
refer
that
to
see
what
configuration
changes,
what
I
will
be
covering
I
will
be
covering
some
configurations
for
you.
I
I
So
you
can
mention
help
check
path
like
health
or
something
like
Ping
that
your
application
is
made
to
responder.
You
can
do
that
over
here.
The
same
goes
for
Readiness
probes.
I
won't
go
into
the
details
of
Readiness
probes
and
Rhymes
Pro,
but
basically
they
are
responsible
to
keep
your
application
Live
and
healthy
in
every
State.
Okay,
so
and
then
we
have
configurations
such
as
Auto
scaling.
I
You
can
mention
the
number
of
minimum
replicas
that
you
want
to
have,
and
then
the
number
of
Maximum
replica
that
what
that
you
want
to
have
these
are
the
replicas
of
your
application.
So
if
the
traffic
on
your
application
increases
and
the
CPU
utilization
suddenly
spikes,
so
this
kubernetes
controller
will
be
automatically
create
more
replicas
of
your
quads
and
it
will
like
redistribute
the
traffic
on
all
the
pods
that
it
has
created.
Okay.
So
this
is
something
that
is
very
useful
in
case
of
I.
I
Don't
know
you
don't
have
a
predictable
traffic
and
there
are
certain
spikes
in
the
traffic
at
your
organization.
So
this
is
pretty
useful.
You
don't
have
to
reserve
instances
and
run
run
your
applications
on
over
reserved
in
census.
Okay,
so
you
can
set
it
at
like
around
80
percent
or
eighty
percent
of
CPU
utilization
and
80
percent
of
memory
utilization.
I
So
as
soon
as
your
application
applications,
average
utilization
goes
above
80
percent,
it
will
create
more
replicas
of
your
application,
and
this
configuration
says
that
as
soon
as
your
configuration
average
memory
utilization
goes
above
80
percent,
it
will
create
more
replicas
of
this
application.
So
either
of
this
condition
hits
it
will
create
more
replicas
of
your
application.
You
can
also
mention
various
other
metrics.
Also
apart
from
just
CPU
and
memory
in
case
of
Kafka
consumers,
it
can
be
the
land
Matrix
in
case
of
your
web
application.
I
It
can
be
the
latency,
it
can
be
the
throughput,
whatever
you
feel
you
can
include
you.
Can
you
can
make
that
Matrix
provide
with
the
HP
and
it
will
Auto
scale
on
the
basis
of
that
particular
Matrix,
so
that
configuration
you
need
to
provide
in
the
extra
position?
Okay,
then
we
have
the
command
section.
I
If
you
don't
have
so,
if
you
haven't
provided
any
command
in
your
Docker
file,
you
can
override
or
if
you
want
to
override
your
command
section
that
you
have
provided
in
the
docker
file.
This
is
the
place
you
can
execute
the
those
commands.
So
this,
if
you
want
to
use
you,
want
to
supply
this
particular
section
and
the
deployment
template
itself.
You
can
do
that,
otherwise
it
will
be
picking
it
from
the
docker
file
itself.
I
I
So
if
you
check
this
application
is
not
available
at
this
a
particular
URL
at
present,
because
we
haven't
deployed
it
yet
so
I
am
just
enabling
this
Ingress
for
this
particular
application
and
then
we'll
be
deploying
this
going
ahead.
Then
you
can
give
resources
for
your
application.
So
these
are
the
CPU
and
memory
resources
that
you
can
allocate
to
your
application.
Containers
I
will
be
keeping
it
a
little
low
just
to
demonstrate
you
how
Auto
scaling
works.
I
So
these
are
so
0.01
is
like
100
of
core
either
you
can
specify
it
in
form
of
ports
or
you
can
also
specify
it
in
forms
of
M,
which
is
mini
CPU.
So
one
one
core
is
equal
to
one
thousand
Milli
CPUs,
so
you
can
do
it
in
either
form.
This
is
in
these.
Both
forms
are
supported
in
kubernetes.
Okay,
so
I
am
giving
it
in
forms
of
four
zero
one
zero
one.
I
And
then,
if
you
have
some
sort
of
volumes
that
you
want
to
mount,
you
can
specify
what
it
over.
Here
you
have
some.
If
you
want
to
create
your
service
as
a
load
balancer,
you
can
specify
it
over
here.
In
this
case,
I
will
be
using
Ingress,
so
I
will
be
using
the
cluster
IP.
So
cluster
IP
is
for
all
the
services
that
should
be
accessible
within
the
cluster.
Then
we
have
an
another
one
which
is
load
balancer
type,
and
then
we
have
a
third
one,
which
is
the
node
portal.
I
So
this
configuration
is
saved.
Let
me
go
to
the
next
step,
which
is
the
workflow
system,
so
I
will
be
making
a
workflow
for
my
environment
so
for
this
workflow
I
will
be
creating
a
CI
pipeline.
So
this
is
what,
where
the
build
will
be
happening,
so
I
will
be
doing
it
from
building
it
from
my
main
branch
and
then
you
can
have
it
automatic
or
manual.
You
can
select
it
as
automatic
or
manual.
I
Then
you
have
certain
premi
build
and
post
build
stages
that
you
can
enable
from
here
in
case
of
free
build
stages.
You
can
download
stuff
that
your
Docker
file
depends
on.
You
can
do
that
in
the
pre-built
stages.
In
the
post
build
stages.
You
can
run
some
sort
of
scans
if
you
want
to
integrate
with
the
third
party
tool.
You
can
also
do
that
over
here.
I
So
this
is
all
well
supported
in
a
different
as
well,
so
I
will
be
creating
this,
so
I
will
be
scanning
my
images
so
for
the
vulnerability.
So
I
have
enabled
this
one,
so
I
have
created
a
CI
pipeline,
so
this
is
my
CI
Pipeline
and
then
I
will
be
attaching
the
CI
pipeline
to
a
CD
pipeline
as
well,
which
is
the
deployment
pipeline
where
I
will
be
deploying
my
code.
So
I
will
be
deploying
my
code
in
the
dev
environment.
I
We
give
you
option
to
select
multiple
deployment
strategies
that
are
not
supported
out
of
the
box
in
kubernetes
as
such.
So
here
you
can
select
among
rolling
blue
green
canvory
and
recreate
I
will
be
creating
a
blueprint
deployment,
so
I'll
set
it
as
movie,
so
in
the
CD
pipeline.
Also,
you
have
option
to
sort
of
Define
your
pre
deployment
stage
or
the
post
deployment
straight.
If
you
can
see
here
over
here
so
in
the
pre-deployment
stage,
you
might
want
to
run
some
sort
of
the
database
migration.
I
I
Okay,
so
I'll
I
have
just
switched
from
the
configuration
mode
to
the
trigger
mode
now,
so
this
is
the
trigger
mode,
so
in
trigger
mode
I
can
select
commit
from
where
I
want
to
build.
I
can
see
which
some
information
about
the
summit.
If
you
don't
see
any
any
of
these
comments,
you
can
Force
refresh
this.
If
you
want
to
do
that,
but
it
mostly
fetches
at
regular
intervals,
so
you
should
be
able
to
see
the
greatest
Comics.
I
So
this
is
my
latest
comment
so
that
I
will
be
building
my
image
from
so
I
will
just
select
it
and
click
on
start
build.
So
so
it
has
started
building
this
particular
it
has.
Building
image
from
it
has
started.
Building
image
from
this
particular
Docker
file
that
I
mentioned.
So
if
you
want
to
see
the
logs
of
these,
you
can
check
that
you
can
make
it
full
screen.
You
can
browse
through
the
logs.
You
can
see.
I
I
So
since
this
particular
pipeline
that
I
created
was
automatic,
so
as
soon
as
it
built
it
finished,
building
the
script
it
has
triggered
this
pipeline,
you
can
see
it
isn't
progressing,
so
it
has
triggered
this
pipeline
automatically,
as
I
have
posted
my
pipeline
to
be
automatic.
So
let's
see
how
the
deployment
is
going
going.
I
I'll
go
to
the
app
Details
page
and
I
I
can
see
that
two
ports
are
up
and
running
I
can
check
the
events
of
these
parts.
So
these
are
the
events,
so
you
can
see
that
this
particular
deployment
has
been
created
and
it
has
been
assigned
to
this
particular
node.
It
has
pulled
these
images
and
it
has
pulled
these
images.
I
That
is
just
word,
so
let
me
come
back
to
this
application,
so
this
is
the
beauty
of
that
from
that,
while
you
don't
have
to
create
any
kubernetes
yamls
or
write
any
kubernetes
yamus,
you
can
still
view
all
the
kubernetes
yamas
that
it
has
created
so
for
this
particular
application
it
has
created
a
replica
set.
It
has
created
a
port.
You
can
check
the
Manifest
of
these
spots,
so
these
are
the
kubernetes
yamls
for
these
parts.
So
it's
it.
It
also
provides
a
lot
of
learning
opportunities
for
the
developer
community.
I
So
if
you
are
not
very
used
to
kubernetes
deployments
and
writing
kubernetes
yamas
or
you
are
not
very
used
to
kubernetes
in
infrastructure
and
environment,
you
can
learn
stuff
from
this
as
well.
So,
while
deploying
your
application,
you
can
see
what
yeah,
what
configuration
or
yaml
devtron
has
created
and
learn
stuff
as
well
from
here.
So
if
you
can
see
I
I
mentioned
the
resources
to
be
0.01
M.
I
So
if
I
go
and
see
my
resources,
so
let
us
given
10m,
which
is
like
one
tenth
of
the
four
right
and
I
mentioned
100
Mi,
so
it
has
given
the
resources
100
Mi.
This
is
the
Docker
image
that
we
just
built
using
the
CI
pipeline
Okay.
So
we
provide
you
all
the
kubernetes
Native
manifest
you
can
select
any
kubernetes
native
manifest
for
this
particular
deployment.
It
has
created
a
pod,
it
has
created
a
service.
Then
it
has
created
an
Ingress
that
I
mentioned.
I
Then
it
has
created
a
service
monitor
as
well.
So
all
the
kubernetes
Native
manifests
are
available
for
you
to
see.
So
if
you
want
to
see
the
auto
scaling
HQ
manifest
I,
have
you
can
see
that
over
here
all
right?
So
this
was
for
so
our
app
has
been
deployed.
Let's
go
back
and
check
on
demo.defron.ai.
I
I
So
if
I
hit
it
one
more
time,
this
Sport
and
this
La,
this
request
is
landed
on
RW,
Port,
okay
and
then
one
more
time
if
I
hit
it,
it
has
the
Android
one
gh7
port,
so
kubernetes
balances
the
load
equally
or
among
all
your
ports.
So
this
feature
is
very
helpful
for
you
and
when
it
comes
to
Auto
scaling
your
infrastructure
or
Auto
scaling
your
application,
so
I'll
quickly
demonstrate
how
it
happens.
I
So
for
that
I'll
have
to
pull
down
this
particular
threshold,
so
I
gave
the
CPU
threshold
as
80.
So
it's
not
going
to
reach
80
percent.
Let's
keep
it
something
around
five
percent
to
see
if
we
are
able
to
hit
this
threshold
and
then
our
application
Auto
scales
or
not.
So
let
me
just
change
this
parameter
and
then
deploy
it.
I
So
these
are
the
monitoring
graphs
that
are
integrated
within
their
front,
so
we
provide
you,
CPU
use
it
memory
usage
throughput
the
2x6
request
that
your
pods
are
getting
the
5xx
if
your
web
application
is
getting
any
the
latency
as
well.
So
these
are
the
monitoring
graphs
that
are
integrated
within
your
different
UI
itself,
foreign.
I
I
I
I
So
if
you
see
the
CPU
utilization
here,
it
is,
it
has
increased,
they
have
receivable
utilization
has
increased
and
it
has
created
more
replicas
because
of
that,
so
that
was
about
application
deployment
using
a
different
you
can
for
some
diagnostic
I
would
like
to
demonstrate
some
diagnostic
troubleshooting
sort
of
functionalities
are
different,
which
has
so
you
can
check
the
application
logs.
So
if
you
click
on
any
what
you
you
have
option
to
check
its
log,
so
these
are
the
application
logs.
I
You
have
functionality
to
be
able
to
grip
on
these
lock.
So
if
I
want
to
grab
I
can
use
this
crap
option,
then
I
can
even
exec
inside
the
container
and
check
if
something
I
have
to
do.
If,
if
as
a
debugging
process,
if
I
have
to
exit
inside
the
container
in
the
running
container,
I
can
also
do
that
using
this
functionality
terminal.
So
I
can.
I
Lifting
big
stick
inside
this
one:
okay,
so
I
have
successfully
exited
inside
this
one.
So
if
I
want
to
run
some
sort
of
commands,
I
can
do
that.
I
can
see
my
application
files
are
present
inside
the
container.
So
this
is
also
very
helpful
when
you
are
trying
to
debug
your
application
and
then
we
also
show
you
all
the
events,
so
all
the
kubernetes,
even
so
on
the
UI
itself,
and
then
you
also
have
the
option
to
see
the
Manifest
as
well.
I
So
if
I
check
this
one,
so
it
shows
me
all
the
vulnerabilities
this
app
has
So.
Based
on
these,
you
can
create
rules
that
you
don't
want
to
permit
any
sort
of
vulnerabilities
any
critical
vulnerabilities.
So
you
can
do
that
as
well.
So
that
was
something
about
it.
So
I
guess
we
have
not.
We
don't
have
enough
time,
so
I'll
just
quickly
get
back
to
slides.
I
So,
like
I
told
you
this,
the
different
project
is
a
totally
open
source
project
and
you
can
contribute
to
the
devtron
project
as
well.
So
recently
we
started
the
sort
of
opportunity
for
all
you
audio,
devops
and
enthusiasts
and
open
source
enthusiasts.
So
you
can
contribute
your
chart
so
recently,
we'll
be
releasing
of
nature,
bring
your
own
chart.
So
there
you
can
bring
your
you
can
create
your
own
chart,
Helm
chart
and
integrate
it
with
devtron.
I
I
On
GitHub
repository
and
inside
this
contribute
chart
chart
folder,
you
will
find
the
specifications,
how
you
can
contribute
your
chart
and
when
we
will
be
releasing
this
feature
of
build
your
own
or
bring
your
own
chart
will
be
adding
these
Community
charts
to
those
section,
along
with
the
proper
credits
given
to
your
name.
So
that's
about
it
I
think
we
we
are
available
at
Discord
for
any
sort
of
installation
help
or
any
sort
of
troubleshooting
that
you're
facing
with
or
their
front.
I
I
I
I
So
if
you
see
we,
we
provide
you
all
the
configurations
right
ahead
itself,
so
beat
replica.
Set
beat
pods
all
sort
of
kubernetes
objects
that
it
creates
it.
It
sort
of
gives
you
that
many
plus
files
for
this.
This
is
not
a
deployment
actually,
so
we
use
rollout
so
in
case
you
want
to
get
the
rollout
manifest
it's
right
here.
This
is
not
the
deployment
kubernetes
object,
so
we
use
rollout
with
devtron
because
of
some
enhanced
features
that
we
have
in
rollout.
I
C
I
Right
so
I'll
quickly
show
you
so
we
have
so
apart
from
this
particular
future
and
contribution
that
I
asked
for
you.
So
so
you
can
contribute
charts
or
you
can
even
contribute
to
features
on
different.
So
for
that
you
can
go
to
the
issue
section.
You
will
see
a
lot
of
issues
section
which
are
already
open
a
lot
of
issues
that
are
already
open.
I
So
if
you
want
to
volunteer
and
work
on
some
issue
that
has
not
already
been
picked
for
by
someone,
you
can
quick,
you
can
just
contribute
to
that.
So
you
can
connect
with
the
team
on
that
issue,
and
you
can
say
that
you
want
to
volunteer
and
contribute
regarding
this
issue
and
we
will
be
more
than
happy
to
have
you
set
up
the
things
for
the
contribution
and
we'll
take
it
and
merge
it
and
release
it
in
the
next
releases.
I
C
Cover
of
davetron,
so
that
would
be
really
helpful
for
all
of
you
to
get
started
with
your
open
source
journey
and
if
you
all,
if
you
are
already
doing
that,
then
it
would
be
a
benefit
for
that
yeah.
Thank
you
so
much
prakash
for
the
great
session
people
really
loved
it
and
your
demo
was
great.
So
thank
you
so
much.
C
C
C
C
C
C
C
B
B
Yeah
I'm
good,
so
so
let
me
introduce
Ben
first,
so
he's
experienced
technologist
with
a
demonstrated
history
of
team
leadership,
Business
Development
and
Community
engagement.
He
skilled
and
confident
Communicator
with
pre-saints
conference
speaking
experience,
historic
professional
with
a
background
in
software
engineering,
solution,
architect
and
business
analysis.
So
it's
it's
a
bit
of
I
can
say
some
like
of
the
words.
But
let
me
tell
you
like
he's
a
very
good
speaker,
because
I
have
previously
also
sent
him
on
other
I
can
say
sessions,
so
you
can
just
use.
B
J
Sure,
thanks
very
much
no
pressure
at
all.
Thanks
for
telling
everybody
that
I'm
a
great
speaker,
it
means
that
I've
got
to
really
nail
it
today.
One
of
the
things
that
I
want
to-
let
you
all
know
about,
though,
is
the
kind
of
presentations
I
like
to
give
are
interactive,
so
I've
got
the
chat
up
here.
Ask
questions
throughout
I'm
going
to
try
and
keep
an
eye
on
that.
If
not,
we
can
cover
questions
at
the
end,
but
please,
if
there's
anything
that
I'm
talking
about
that.
J
You
want
me
to
go
over
in
more
detail
if
you've
got
any
questions
that
are
relevant
to
what
I'm
talking
about
at
the
time
ask
them
we
can
go
through
it.
My
Talk
itself
is
probably
usually
only
about
half
an
hour,
but
we've
got
40
minutes,
I.
Think
45
minutes.
Maybe
to
do
it
today,
which
means
we've
got
plenty
of
time
to
really
get
into
the
nitty-gritty
about
identity,
which
is
what
the
talk
is
about.
So,
as
has
already
been
mentioned,
my
name
is
Ben
deckroy
I'm,
a
developer
advocate
for
auth
zero.
J
All
zero
are
essentially
like
a
software
as
a
service
platform
for
identity,
so
rather
than
writing
your
own
login
and
your
own
password
reset
and
single
sign-on
and
Enterprise
logins
and
all
of
these
things.
Auth
zero
is
a
product.
In
the
same
way
as
you
wouldn't
write,
your
own
credit
card
payment
stuff,
you
would
use
stripe
or
PayPal
or
or
some
other
mechanism
or
zero
is
the
kind
of
product
you
can
use
for
all
of
your
identity.
J
J
That's
not
actually
my
name!
So
when
we
talk
about
identity,
we're
talking
about
identifying
who
was
logged
on
based
on
previous
credentials,
so
the
next
time
they
log
on.
We
know
it's
the
same
person,
whoever
that
person
might
be
okay,
let's
get
into
it.
So
click
on
my
presentation,
so
I
can
actually
actually
works
so
where,
where
did
identity
all
start-
and
this
is
the
first
part
of
the
the
interactiveness
I'd
like
you
to
like-
put
a
hands
up
emoji
or
a
plus
one
or
something
in
the
chat.
J
J
What
we'll
do
is
we'll
make
it
easier
for
people
to
do
this,
so
what
they
would
do
is
they'd.
Have
this
friend
finder
Facebook
called
the
Friend
Finder?
There
were
various
ones
for
different
systems
and
they
would
say
tell
you
what,
if
you
give
us
your
email
address,
and
then
you
tell
us
the
password
that
you
use
to
log
into
your
email.
Then
we
can
log
into
your
email
account
for
you
and
we'll
have
a
look
at
your
contact
list
and
we
will
email
all
of
your
contacts
and
invite
them
to
Facebook.
J
It's
a
brilliant
idea,
sort
of
and
I'm
sure
a
lot
of
you
are
already
going.
Oh
cringe
that
that's
terrible,
who
even
thought
of
this,
and
some
of
you
might
be
thinking.
Well,
that
seems
reasonable.
Why
wouldn't
you
do
this
and
we'll
cover
that
one
of
the
things
about
software
development
is
that
we
don't
all
have
the
time
to
focus
on
everything
security
wise.
So,
if
you're
sitting
there
thinking,
you
know
what
I
don't
actually
know
what
the
issue
is,
there's
nothing
wrong
with
that.
J
J
So,
yes,
the
the
mechanism
would
log
in
and
and
and
get
off
all
of
your
email
addresses.
So
this
is
roughly
how
it
worked.
You
have
the
the
user.
On
the
left
hand,
let
me
use
my
mouse
and
make
it
easier
for
you
to
see
what
I'm
looking
at
to
see
for
the
user
over
here
and
they've
got
their
credentials.
They've
got
their
username
and
password
for
logging
into
their
email
account
and
over
here
there's
some
kind
of
social
network.
J
Let's,
let's
just
say
Facebook
for
now
it
could
have
been
many
of
them.
I'm,
not
attacking
Facebook
lots
of
different
systems.
Did
it
and
the
reason
they
did
it
was
there
was
not
really
a
better
way
of
doing
it.
What
they
would
do
is
say:
look
we
want
you
to
provide
us
with
your
email,
account
credentials
and
then
oops
I'm
clicking
the
wrong
button
here.
Go
in
this
direction.
J
Here
there
we
go,
provide
us
with
your
email,
account
credentials,
and
then
what
we'll
do
is
we'll
send
those
credentials
to
your
email
account
provider,
so
that
could
have
been
Yahoo
mail
or
Gmail
or
Hotmail
or
any
of
these
kind
of
email
providers.
We
will
log
in
as
you
so
this
isn't
using
an
API
or
any
kind
of
standard
mechanism
for
third-party
systems
to
talk
for
each
other.
They
were
essentially
pretending
to
be
you,
logging,
in
insofar
as
Hotmail
or
whoever
didn't
even
know
that
it
was
a
server
or
an
automated
script.
J
It
is
hard
and
Facebook
don't
want
to
have
that
problem,
or
they
didn't
want
to
have
that
problem.
They
have
it
now,
so
what
they
would
do
is
they
would
pretend
to
be
this
human,
you
and
tell
Hotmail
or
Yahoo
mail
to
send
the
email
on
your
behalf,
which
means
that
your
friend
gets
an
email
from
you
not
from
Facebook
or
LinkedIn,
but
also
it
came
from
the
Hotmail
server,
which
means
deliverability
is
higher
and
the
likelihood
of
getting
caught
in
spam
is
lower.
J
So
that
was
a
great
idea
in
theory,
but
it's
also
a
little
bit
dubious
in
terms
of
using
Hotmail
services
or
whichever
email,
Provider
Services
for
this
purpose,
when
it's
not
even
their
service,
it's
like
the
social
media
provider
is,
to
a
certain
extent,
possibly
misusing
those
mail
services.
So
there
were
ethical
issues
with
this,
but
but
it
was
a
the
simplest
solution
of
the
time
and
then,
as
I
say,
they
would
take
all
those
those
contacts
and
they
were
put
into
the
database.
So
as
well
as
having
emailed
all
your
friends.
J
They've
now
got
your
friends
email
addresses
in
the
database
so
that
when
their
friends
then
join,
they
can
make
that
link
straight
away
on
a
slight
tangent,
and
this
has
nothing
to
do
with
Identity
or
authentication.
Another
thing
that
social
media
providers
do
is
a
lot
of
them.
Do
I'm
not
going
to
accuse
all
of
them
of
doing
it,
but
a
lot
of
them
will
do
it.
You
might
have
heard
about
Shadow
accounts
or
Shadow
profiles
and
Facebook
I'm.
Pretty
certain.
J
Does
this,
for
example,
where
if,
if
you
use
this
system
and
my
email
address,
got
added
into
Facebook
and
I
never
joined
Facebook,
it
would
have
this
Shadow
dark
profile.
There's
there's
been
in
person,
we
don't
know
much
about
him.
He
hasn't
joined
our
system,
but
we
know
that
he's
friends
with
this
person
over
here,
because
that's
how
they
joined
in
and
over
time,
they
might
start
building
a
profile
of
where
I
might
have
been,
depending
on.
J
So
this
was
this
was
the
major
issue
this
big
exclamation
mark
over
here
provided
me
providing
my
email
account
credentials
for
logging
into
my
email
to
a
third-party
system.
They
can
then
log
into
my
account
and
check
my
contacts,
but
they
can
also
read
all
my
email
send
email
as
me.
They
could
change
my
account
settings
and
potentially
they
could
even
go
in
and
start
looking
at
my
billing
settings
and
see
what
the
last
four
digits
of
my
credit
card
and
like
they
had
full
access-
and
this
was
a
major
issue.
J
I
was
basically
providing
somebody
access
as
though
they
were
me,
so
this
obviously
wasn't
ideal
and
oauth
oauth
2.0,
specifically
there's
another
author
of
one,
which
is
very
very
different,
and
we're
not
talking
about
that
today.
At
all
you,
you
won't
see
it
used
very
much
if
at
all
and
and
it
has
a
different
solves,
a
different
problem,
so
oauth
2.0
is
basically
an
authorization
mechanism
for
authorizing
systems
to
talk
to
each
other.
So
this
is
how
oauth
works.
J
J
Your
email
provider
is
then
going
to
generate
some
kind
of
token
of
some
sort,
and
it's
going
to
return
that
through
a
redirect
back
to
the
server
making
a
request
for
the
information.
So
this
token
is
actually
like
an
authorization
code.
It
doesn't
really
have
anything
inside
it,
that's
particular
to
to
you.
It
doesn't
have
any
identity
information
in
it
or
anything
like
that,
but
this
token
can
be
then
sent
from
the
social
media
provider
directly
to
the
mail
provider
via
a
standard
API.
J
So
we've
got
apis
in
play
now
and
it
says:
look
some
user
that
that
you've
just
issued
this
token
to
wants
to
give
me
access
to
their
contact
list,
and
this
token
will
actually
have
inside
it
a
set
of
Scopes
and
the
scope.
One
of
the
Scopes
will
be
access
the
contact
list.
So
that's
the
only
thing
they
can
do.
They
can
receive
their
contact
list
back.
They
can't
send
emails,
they
can't
read
emails.
They
can't
find
out
the
last
four
digits
of
my
credit
card
number.
J
This
is
a
very
tightly
scoped
a
tightly
specified
purpose.
For
this
token,
they
can
only
get
the
contact
details
so
now
they
need
to
find
their
own
way
of
doing
email,
and
this
is
the
right
way
of
doing
it.
They
can
either
sort
the
problem
out
themselves
or
in
the
same
way
as
auth.
Zero
is
a
software
as
a
service
provider
for
identity.
There
are
plenty
of
options
for
sending
emails
through
apis
as
well,
so
they
can.
J
J
The
next
thing
we
want
to
look
at,
or
let's
just
have
a
just
to
kind
of
ground.
Exactly
what
authors
of
the
permission
system,
some
examples
might
be
like
read:
LinkedIn
can
read
emails
on
Gmail.
You
might
not
want
to
be
able
to
write
emails,
but
you
have
these
three
things
that
the
the
actor
can
take
an
action
on
a
Target.
So
that's
what
authorization
is
another
example
might
be
tweetdeck
and
post
tweets
on
Twitter
tweetdeck
wasn't
originally
owned
by
Twitter.
J
But
tweetdeck
then
has
permission
to
take
actions
on
your
behalf
on
Twitter
and
if
Enterprise
is
another
one,
whenever
I've
created
events
in
the
past
on
Eventbrite
I've
actually
connected
that
to
Facebook
accounts
associated
with
the
events
that
I'm
running
and
Eventbrite
can
get
permissions
to
create
an
event
on
Facebook
directly.
So
I
don't
need
to
do
it
in
two
places:
I
create
an
event,
an
Eventbrite
I
click
on
push
to
Facebook,
and
it
sorts
all
of
that
out.
J
So
these
are
some
kind
of
examples
of
authorization,
there's
not
authentication,
which
is
finding
out
who
the
user
is
actually
I.
Just
asked
can
oauth
access
your
info,
like
Facebook,
can
so
we'll
get
on
to
that
in
just
a
second,
but
just
to
let
you
know
that,
like
Facebook
I'm,
not
sure
what
you
mean
by
like
Facebook
can
because
when
you,
let's
just
go
on
to
the
next
few
slides
and
maybe
I'll
answer
your
question,
so
here
it
is,
Eventbrite
can
get
Ben's
identity
from
Twitter.
J
J
So
this
is
where
user
info
comes
in
and
essentially
user
info
is
a
layer
that
sits
on
top
it's
it's
almost
like
another
API
endpoint,
it's
a
layer
that
sits
on
top
of
the
the
oauth
protocol
and
it
was
designed
to
solve
this
problem
of
finding
out
who
the
user
in
another
system
was.
This
is
the
basically
like
the
login
with
Facebook
type
thing,
so,
once
a
once,
two
systems
have
have
this
authorization
permissions
between
them.
One
of
the
actions
can
be
get
user
info.
J
J
In
their
token,
so
what
get
user
info
does
is
it
creates
another
endpoint
now
right
at
the
beginning,
when
this
was
first
streamed
up,
it
was
basically
a
an
inconsistent
mechanism
across
the
different
providers,
so
one
provider
for
for
user
info
might
just
be
slash.
User
info,
one
might
be
slash
user,
slash
info,
one
might
be
something
completely
different,
one
might
be
versioned,
one
might
not
be
and
you'd
pass
in
a
bearer
token
again,
because,
obviously
you
need
to
prove
that
you
have
the
scope
or
the
permission
to
read
the
user
information.
J
But
then
the
response
that
came
back
could
be
anything
if
it's
is
it
XML
or
is
it
Json
or
is
it
yaml
or
is
it
plain
text
or
like
there
was
there
were
no
standards
around
this
different?
If
you,
if
you
had
two
or
three
already
logging
with
Facebook
and
login
with
Google,
and
you
wanted
to
add
login
with
Twitter,
then
you
might
have
to
read
the
separate
specs
and
develop
a
slightly
different
mechanism
for
adding
another
provider
for
login
by
using
this
user
info
endpoint.
J
So
there
was
no
consistency
around
it,
but
it
was
definitely
another
step
forward.
It
was.
It
was
a
consolidation
of
an
idea
that,
yes,
we
need
to
provide
some
kind
of
user
info
endpoint.
So
what
happened?
Was
the
industry
came
together
and
said?
Hey
we've
got
like
75
billion
different
implementations
of
this
user
info
thing.
Let's
create
one
standard,
we'll
call
it
open.
Id
connect
now
in
the
same
way
as
oauth,
one
and
oauth
2
are
very
different.
There's
open,
ID
and
open
ID
connect
also
very
different.
J
So
if
you've
heard
of
open
ID
again
we're
not
going
to
talk
about
that
today,
but
open
ID
connect
itself
is,
if
you
imagine
this,
this
diagram
with
a
big
box
and
on
it
just
says,
oauth
2.0,
on
top
of
it,
is
open.
Id
connect,
openid
connect
is
dependent
on
all
2.0,
because
it's
that
user
info
extension,
but
it's
a
standards,
compliant
mechanism
for
getting
that
user
info.
J
J
It
is
used
to
verify
identity
so
once
I'm
logged
into
Twitter,
for
example,
if
I
go
to
the
user
info
endpoint.
With
that
token,
it's
going
to
verify
what
Twitter
knows
about
me
so
that
you
can
do
login
with
Twitter
and
get
that
information
back,
and
this
goes
back
to
perhaps
action's
question
before
about
getting
identity
from
from
Facebook
log
on
with
Facebook
would
use
this
simple
identity
layer
on
top
of
oauth
2,
to
talk
to
Facebook
and
get
identity
information
back.
J
It
contains
not
just
the
the
identity
in
terms
of
an
ID
number
or
something,
but
it'll
actually
have
basic
profile
information.
Maybe
it's
got
your
email
address,
might
have
your
name,
possibly
a
URL
for
a
profile
photo,
and
it's
always
an
interoperable
and
rest
like
manner
so
to
make
it
consistent
and
easy
soap
is,
is
like
a
protocol
that
was
used
before
rest
have
come
in
and
we
thought
you
know
what,
let's,
let's
keep
it
really
consistent
and
modern
and
up-to-date
so
rest
like
Manner
and
finally,
the
Json.
J
The
data
format
that
comes
back
is
in
Json,
and
the
beauty
of
this
is
that
you
can
have
structured
data
in
there.
You
can
have
nested
information,
you
can
have
arrays
of
information,
so
it
really
allows
you
as
a
a
creator
of
a
profile
information
if
you
are
the
Facebook
or
the
the
Twitter
or
the
Googles
of
the
world.
J
If,
if
you
allow
people
to
log
in
through
your
site,
then
you
can
provide
information
back
in
a
way
that
whoever's
consuming
is
always
going
to
know
it's
Jason,
but
the
beauty
is
that
you
can
put
whatever
you
like
in
there
all
right.
So
how
does
openid
connect
work?
You
can
see.
We've
got
an
extra
element
in
here
now
we've
got
the
identity
provider
down
at
where's,
my
mouse.
There
is
identity
provider
down
at
the
bottom.
So
this
could
be
all
zero.
J
It
could
be
an
off-the-shelf
product,
it
could
be
whatever
it
is:
don't
roll
your
own
in
the
same
ways
you
don't
roll
your
own
credit
card
payments,
don't
roll
your
own
identity
and
login?
It's
like
there's
so
much
that
goes
into
it.
You
don't
want
to
get
that
wrong.
This
is
your
server
up
here,
somebody's
trying
to
to
make
a
request
for
a
resource
and
there's
a
login
page.
That
appears
from
somewhere,
let's
see
how
all
these
top
bits
tied
together.
So
a
user
comes
along,
oh
yeah.
J
J
So
this
is
the
the
client
relationship
we're
talking
about
here.
Not
the
client
has
a
new
user,
but
the
client
has
in
your
web
application.
So
you'll
have
a
client
ID,
which
is
this
little
person
face,
because
it's
not
like
an
ID
and
a
client
secret.
That's
the
key!
It's
a
secret!
Essentially
this
key
or
this
climb
secret
is
only
ever
used
only
ever
sent
back
to
the
identity
provider
to
prove
that
this
client
is
the
one
that's
authorized
to
talk
to
the
audience
provider
about
about
this
whole
transaction.
J
So
user
comes
along
and
makes
a
request
for
some
kind
of
resource
from
your
server.
Maybe
it's
a
web
page
they're
trying
to
get
to
a
dashboard
or
maybe
they're,
trying
to
download
a
PDF
or
or
maybe
they're
trying
to
take
an
action
like
make
a
request
to
to
to
update
some
information
or
something
and
whereas
before
they
were
allowed
to
see
things
but
to
post
there
needs
to
be
some
kind
of
identification
of
who
the
user
is.
J
So
your
application
is
going
to
say
here:
I,
don't
know
who
you
are
we're
going
to
start
a
login
process
and
what
it'll
do
is
it'll
return,
a
redirect
request.
So
this
part
on
the
left
here
is
kind
of
like
a
browser
as
well,
so
send
a
redirect
request
to
redirect
the
user
to
the
identity
provider,
and
you
can
see
there
that
the
the
ID
the
client
ID
was
passed
along
in
there
too
there's
a
couple
of
other
bits
of
information
there
like
Scopes.
J
What
do
you
actually
want
to
know
about
the
user
and
open
ID
would
be
one
of
those
Scopes,
because
we
want
to
know
about
the
profile.
A
couple
of
other
Scopes
and
and
the
identity
provider
can
then
use
this
client
ID
and
say:
hey.
I
know
this
is
John's
website,
so
we're
going
to
ask
this
user
to
log
into
John's
website,
so
we'll
use
the
database
of
users
that
are
on
John's
website.
J
J
J
J
It's
a
very
secure
way
of
doing
that
and
the
credentials
are
then
sent
back
to
the
identity
provider.
Again,
you
don't
need
to
worry
about
storing
passwords
and
hashing
algorithms
and
password
resets
and
all
of
these
kind
of
things,
and
then,
if
you
notice,
just
in
here
once
you've
logged
in
the
identity
provider,
is
going
to
convert
that
into
some
kind
of
user,
says:
okay.
Well,
we
know
that
Ben's
just
logged
in
and
we're
going
to
send
this
we're
going
to
generate
this
authorization
code,
which
is
a
random
string.
J
J
So
if
we
just
look
at
the
flow
so
far,
I've
made
a
request
to
your
website.
Your
website
says
here
I'm,
going
to
redirect
you
to
the
login
server.
Login
server
gives
me
a
login
page
I
provide
my
credentials
to
the
login
server
login
server.
Then
redirects
me
back
to
your
web
application
and
your
web
application
now.
Has
this
authorization
code
doesn't
say
anything
about
me
though,
but
what
I
can
do
is
I
can
take
this
authorization
code
and
the
client
secret,
and
you
can
send
that
to
the
identity
provider
directly.
J
J
Those
tokens
are
Json
web
tokens,
and
this
is
what
a
Json
web
token
looks
like.
Does
that
make
sense
all
right?
Let's
move
on?
No,
let's,
let's,
let's
dive
into
this,
a
little
bit
more.
So
if
you
haven't
seen
an
access
token
before
it
might
look
a
little
bit
overwhelming,
but
essentially
it's
base64
URL
encoded
data.
Let's
just
color
code,
it
a
little
so
we
can
see.
What's
going
on,
we've
got
a
header,
that's
the
the
red
part
and
then
in
the
verbally
color,
the
middle
bit.
That's
the
payload!
J
This
is
the
data
about
me
and
then
the
third
part
is
the
signature,
and
you
can
see
the
two
highlighted
full
stops
there,
those
those
are
what
I
used
to
separated.
So
when
an
access,
token
or
an
ID
token
or
any
kind
of
Json
web
token
comes
in,
we
can
split
it
on
those
dots
and
there
are
other
formats
of
Json
web
tokens
as
well.
J
So
you
don't
need
to
worry
about
that,
but
it's
useful
to
understand
under
the
hood.
What's
what's
happening,
so
we've
got
ahead
of
the
payload
and
the
the
footer,
the
header.
If
we
break
it
down,
if
we
base
64
Euro
encode
it
it's
basically
Json
it'll
be
formatted
slightly
nicer
so
that
our
brains
can
read
it.
We
can
see
that
there's
an
algorithm
and
a
type
there's
a
couple
of
other
things
in
there
as
well,
and
the
type
is
Json
web
token,
because
there
can
be
multiple
different
types
of
token.
J
There
are
plans,
possibly
to
add
in
the
same
ways
you
can
have
content
types
that
have
pluses
in
them
application
or
slashes.
Even
like
application
Json,
you
might
have
failed
to
have
a
JWT,
Identity
or
JWT
plus
access
or
something
just
so
that
we've
got
a
better
idea
of
what
the
purpose
of
the
token
is.
But
you
don't
need
to
worry
about
this
right
now.
Essentially,
we
can
just
confirm
here
that
the
type
of
token
is
just
a
standard
Json
web
token.
J
The
algorithm
is
an
important
part
in
this
case
we're
using
an
hs256
or
a
hmx,
sharp
256,
hashing
algorithm.
You
don't
need
to
know
about
hashing,
algorithms
or
any
of
the
slot,
but
just
know
that
to
generate
the
signature
for
verifying
this,
we'll
need
to
know
that
bit
of
information
and
then
the
bottom
part
is
the
identity
part
of
the
payload.
So
you
can
see
SUV.
J
This
is
a
very
common
standard,
key
value
in
the
payload
for
a
Json
web
token,
and
it
stands
for
subject
so
remember
before
when
we
saw
in
the
diagram
that
you
had
your
application
and
the
username
and
password
was
stored
in
the
identity
provider.
Your
application
is
still
going
to
have
a
database
of
whatever
it
is
you're
providing
access
to
for
your
users
and
they'll
still
need
to
be
a
user
model
or
user
table
in
that
database,
but
instead
of
having
username
and
password
in
there,
you'll
just
have
a
subject.
J
So
now,
every
time
I
log
into
your
web
application,
you're
always
going
to
get
the
same
subject.
This
value
here
will
come
through
even
if
I
change,
my
username
or
my
email
address
or
I
log
in
from
a
different
computer
or
anything
changes
that
subject
never
changes,
that's
your
unique
ID
of
who
is
just
logged
in
so
you
can
now
create
a
record
in
your
user
database.
Saying
okay,
create
a
record
with
this
subject
and
here's
some
information
about
him
or
her
that
we
want
to
remember.
J
Have
you
got
name
and
job
title
and
region?
These
are
kind
of
wreck
as
it's
Json
you
can
put
whatever
you
want
in
and
then
there
are
some
other
standard
types
at
the
bottom.
Iat
issued
at
exp
expired.
So
there's
there's
extra
bits
of
information
that
come
out
from
there
and
here's
where
the
signature
comes
in
useful.
So
this
screen
is
getting
a
little
bit
busy,
but
let's
have
a
look
what's
going
on
here.
So
this
part
here
is
what
we
just
saw
on
the
previous
screen.
J
So
this
is
the
header
and
the
payload,
and
essentially,
what
we're
doing
is
we're
taking
a
track.
Shaft
256,
which
is
the
hashing
algorithm
to
create
a
signature,
and
this
hashing
algorithm
requires
a
a
password
or
appreciated
key
of
some
sort,
and
the
result
of
that
is
then
base
64,
Euro,
encoded
and
put
into
this
value
called
Signature.
And
now,
if
we
assert
the
signature
we
just
generated
is
the
same
as
a
signature
that
came
through
in
the
Json
web
token.
Then
we
know
that
the
Json
web
token
hasn't
been
tampered
with.
J
And
again
you
don't
need
to
worry
about
the
implementation
of
this.
This
will
come
out
in
the
sdks
that
you're,
using
for
for
whichever
platform
your
identity
provider
you're
using.
But
it's
useful
to
know
that
this
Json
web
token
once
you've
received
it.
Your
application
can
verify
that
it
hasn't
been
modified
since
the
issuer
created
it
without
having
to
ask
the
issuer.
J
J
J
So
some
other
things
that
you
can
get
are
access
tokens.
So
we
looked
at
the
the
ID
token
just
now.
Here's
an
example
of
an
access
token
and,
like
I,
said
before
the
they're,
often
three
characters.
There
are
a
whole
lot
of
standard
and
they're
not
enforced,
like
you
could
use
ISS
for
something
else,
but
it's
highly
recommended
that
you
don't
because
you'll
confuse
somebody,
but
there
are
a
whole
lot
of
standard,
suggested
key
value.
Pairs.
Iss
is
the
issuer
who
created
this
token.
Seb
is
a
subject
who
was
a
created,
four
who's
who's.
J
The
person
who's
logged
in
AUD
is
the
audience.
This
is
a
really
useful
one
for
apis.
The
API
is
the
audience,
and
now
the
API
receiving
the
access
token
can
say.
Was
this
token
generated
for
me
if
I'm
not
api.example.com,
then
I
ignore
this
token,
because
it
wasn't
generated
for
me
azp
is
the
I
can
never
remember
what
it
stands
for,
but
essentially
that's
the
client
ID,
so
I
now
know
which
application
you
logged
into
you've
got
the
issue.
Debt
you've
got
expired
and
here's
where
the
Scopes
come
in
useful
as
well.
J
J
J
If
you
can
base64
URL
decode
in
your
head,
then
you
could
read
a
Json
web
token.
Anybody
can
read
a
Json
web
token,
so
they're
not
encrypted
what
they
are
done.
What
they
have
done,
though,
is
the
third
part
is
a
signature.
So
it's
taken
the
header
and
the
payload,
and
it's
created
a
signature.
So
you
can
verify
that
later
that
the
payload
has
not
been
tampered
or
modified.
J
J
So
where
is
beyond
well
there's
a
slightly
more
complex
version
of
of
what
we
were
looking
at
before.
But
let's
have
a
look
what's
going
on
here:
we've
got
our
identity
provider
over
here,
and
we've
got
our
our
user
and
our
user
might
be
logging
in
via
web
application
or
via
a
mobile
phone,
and
the
mobile
phone
might
need
to
talk
to
an
API
and
then
you've
got
your
web
application
server
over
here.
J
The
token
is
valid
before
responding.
The
API
can
also
receive
that
token
directly
from
the
server
or
even
via
this
request
here
from
the
mobile
phone,
the
mobile
phone
for
the
Orange
Lines.
These
are
the
login
lines.
The
mobile
phone
and
the
web
application
only
ever
provide
credentials
to
the
open,
ID
compliance
server,
but
the
token
the
mechanism
of
using
Json
web
tokens.
J
It
really
makes
it
powerful
in
terms
of
conveying
this
identity
and
when
you're,
looking
at
access
tokens
as
well,
the
the
authorization
permissions
between
these
applications
and
even
though
it
looks
complicated,
it
simplifies
development.
So
much
so
I
hope
that
that's
been
easy
to
follow.
I've
tried
to
take
my
time.
I've
tried
to
keep
an
eye
on
the
the
questions
as
well.
I
see,
there's,
there's
a
question
from
any
cat
about
compliance
with
md5
or
Shah.
J
J
B
B
B
A
J
B
B
B
A
A
A
B
B
B
People
are
there
and
they
are
just
participating
yeah.
So
we
have
the
previous
quiz
winner
also
with
us,
so
you
can
check
the
names
and
screen.
So
there
was
a
quiz
before
the
session
and
we
are
having
the
names
on
the
screen.
So
if
you
any
one
of
you
win
that,
and
so
you
can
just
ping
us
on
mail,
so
we'll
drop
our
mail
ID
just
fingers
on
mail,
and
we
will
reach
out
to
you
for
sure
yeah
thanks
a
lot
Ben
for
joining
in
have
a
great
day.
C
Thank
you
Aditya
and
thank
you
Ben
hi
pratik.
Let
me
introduce
you
for
to
the
audience.
Okay,
so
I
know
you
all
know
him
by
his
profile.
I
know.
I,
definitely
know
this.
Okay,
so
pratik
is
a
popular
programming
instructor
and
an
a
software
engineer
currently
working
as
with
Google
as
an
sde
3
and
founded
coding
minutes
to
bring
affordable
quality
coding
courses
on
udemy.
He
is
known
for
his
amazingly
simplified
explanation
that
make
everyone
fall
in
love
with
programming.
C
He
has
over
five
years
of
experience
in
teaching
and
has
trained
50
000
students.
That's
a
lot.
His
interactive
resume
on
pratiknarang.com
is
quite
popular.
You
definitely
should
check
that
out.
It's
pretty
amazing!
You
can
get
lots
of
insights.
How
you
can
modify
yours
in
the
past.
He
has
won
many
national
level
hackathons
and
been
to
ACM
ICPC
and
worked
with
SanDisk
too
hello
pratik.
How
are
you.
H
So
I
know
it's
a
it's
a
tricky
topic
generally
people
say
competitive
programming
versus
software
development
I
have
been
invited
to
speak
for
competitive
programming
and
software
development
Okay,
so
I
have
to
justify
both
and
that's
why
I
am
here,
so
it
would
be
short
session.
I
will
not
take
a
lot
of
your
time,
but
if
you
have
any
comments,
any
questions
you
can
reach
me
out
in
the
comments
I'm
just
going
through
them
as
well:
hi
sriniket.
H
So
it's
a
topic
that
a
lot
of
people
get
confused
during
the
college,
whether
to
go
into
Android
web
or
competitive.
So
we
will
try
to
clear
out
some
confusion
today
about
some
myths
about
competitive
and
software
development.
Okay,
so
you
already
had
my
introduction:
I
am
currently
sd3
at
Google,
Hyderabad
and
I,
created
codingminutes.com,
where
you
can
find
a
lot
of
pocket
friendly,
affordable
programming
courses
which
are
hosted
on
udemy
and
yeah.
H
So
that's
what
I'm
working
on
for
the
last
few
months
and
in
two
three
days,
I'm,
bringing
a
very
amazing
course
on
data
structures
for
beginners
and
intermediate
programmers.
My
last
course,
which
I
launched
about
two
months
back
got
a
huge
response
from
100
plus
countries
and
I'm
sure
a
lot
of
you
might
have
also
joined
that
particular
course
as
well.
H
So
yeah
that's
little
bit
about
me
and
more
or
less
you
yeah,
so
for
today's
agenda,
I
will
be
talking
about
what
is
competitive
programming
where
it
is
used
in
software
development.
What
should
a
college
student
do?
Should
he
do
competitive
programming
throughout
all
the
four
years,
or
should
he
do
development
as
well?
H
And
what
are
the
current
hiring
Trends,
how
they
are
changing?
Okay.
So
let
me
start
with
the
fact
that
programming
is
a
skill,
and
all
of
you
want
to
master
that
particular
skill,
and
this
skill
has
different
levels:
okay,
so
competitive
programming.
Generally,
some
people
confuse
it
with
basic
data
structures.
Algorithms,
some
people
confuse
it
with
simple,
simply
solving
problems
on
online
platforms
like
code
Chef
code
forces,
so
basically,
in
my
opinion,
I
rate
this
competitive
programming
as
a
extreme
level
of
programming.
H
It
is
also
known
as
Sports
sport
programming,
where
people
compete
with
each
other
and
the
right,
optimized
algorithms,
so
that
they
are
able
to
get
the
same
results
in
a
lesser
time.
Complexity,
using
a
lesser
memory
as
well.
So
basically
I
do
I,
say:
okay,
if,
if
you're
learning
the
programming
fundamentals
such
as
Loops
functions,
arrays,
you
are
on
the
level
one
where
we
are
talking
about
the
basics
of
programming,
that
I
don't
call
it
as
a
program
competitive
programming.
The
next
level
is
like
you
are
first,
you
learn
the
basic
components
of
programming.
H
H
Most
people
would
need
to
learn
a
bike
or
a
scooter
or
a
active
or
a
car,
to
drive
on
the
road,
but
there
would
be
very
few
people
who
would
also
learn,
though,
that
skill
for
a
F1
race,
maybe
driving
a
car
in
a
F1
race,
is
a
hold
together
different
skill.
It
requires
a
lot
of
effort
to
learn
how
to
drive
in
F1
race,
okay.
So
this
is
what
a
competitive
programming
is.
This
is
about
competing
with
other
people.
H
This
is
about
building
the
most
optimized
algorithms,
but
this
do
you
think,
is
this
driving
in
a
F1
race.
Skill
is
useful
in
real
life,
not
a
lot
okay,
it
can
make
you
a
good
driver
on
the
road,
but
you
might
not
be
using
those
Advanced
algorithms,
optimization
techniques
in
solving
your
day-to-day
life
problems.
Okay,
so
for
most
of
you,
I
would
recommend
that
you
learn
the
basics.
You
learn
the
data
structures,
you
learn
the
algorithms
and
you
reach
this
second
category.
H
Okay,
for
this
one,
those
who
are
passionate
about
those
who
are
actually
want
to
compete
and
yeah,
so
people
are
saying
I,
like
prithik's
interactivism
yeah.
Thank
you.
Thank
you,
so
competitive
programming
takes
the
these
algorithmic
skills
to
the
whole
new
level,
and
there
are
a
lot
of
algorithms.
You
may
not
use
in
your
daily
life,
but
learning
those
algorithms
will
definitely
make
you
a
good
programmer
as
well.
Okay.
H
So
that's
that's
about
competitive
programming,
so
I
found
these
pictures
on
my
Instagram,
so
just
thought
to
put
them
stitch
them
together,
so
that
I
could
make
this
meme,
or
example,
to
explain
you
the
difference.
Okay,
so
this
third
category
I
would
say
it's
like
the
competitive
programming
category,
so
I
hope,
I'm,
making
sense
and
foreign.
H
Yeah,
okay,
so
this
is
level
one.
This
is
level
two,
and
this
is
level
three
okay,
so
this
is
CP
okay,
so
that's
what
I
was
trying
to
tell
you?
Okay,
I
have
a
question
for
you.
All
software
applications
have
one
thing
in
common,
so
who
will
tell
me
what?
What
are
the
things?
What
is
one
thing
that
binds
together
all
software
applications?
Let's
talk
about
zomato
or
Uber,
or
swiggy
or
Facebook
or
any
other
application?
What
do
you
think?
H
H
Tell
me
the
one
thing
that
is
common
in
all
software
applications.
That's
great
people
are
saying
data
and
data
structures.
Yes,
so
all
software
applications
have
one
thing
in
common:
they
work
on
data.
Okay,
talk
about
any
software
application;
it
is
doing
some
kind
of
processing
of
data.
Okay,
so
I
will
give
you
three
examples
and
in
order
to
store
that
data
in
order
to
work
on
that
particular
data,
you
need
some
underlying
storage
as
well.
Okay,
so
data
needs
storage
and
storage.
H
H
And
for
Ram
we
have
what,
in
order
to
store
the
data
in
the
ram,
we
have
data
structures.
Okay,
and
even
these
databases
are
inspired
from
the
concept
of
data
structures.
You
might
have
heard
about
B
plus
trees,
okay,
so
they
they
are
used
in
creating
databases
you
might
have
heard
about
graphs,
algorithm
or
graph
algorithms.
So
graph
is
a
data
structure
and
there
are
also
databases
which
are
inspired
from
the
concept
of
graphs
as
well.
H
Okay,
so
all
this
data
that
you're
storing
either
in
the
disk
or
in
the
ram
it
is
stored
in
data
structures
or
in
databases.
Okay,
so
Ram
has
limited
memory.
Okay,
it.
It
only
stores
the
data
that
is
currently
required
for
the
processing,
but
disk
is
like
a
SSD
or
a
hard
disk.
Okay,
that
stores
all
the
data
that
so
this
data
will
persist
permanently.
Okay,
so
this
data
is
persisting
permanently
in
the
data.
Centers
and
RAM
data
is
persisting
for
a
shorter
duration.
H
Maybe
the
fresh
data
is
in
the
ram
after
it
has
been
processed
and
it
has
been
moved
to
SSD
okay,
so
the
data
goes
from
Ram
into
SSD
and
it
maybe
it
comes
from
disk
into
RAM
as
well.
So
it
all
depends
upon
the
application
logic
which
data
to
keep
in
Ram,
which
data
to
keep
in
disk,
and
sometimes
we
also
have
a
memory
which
is
known
as
cache.
H
H
There
are
SQL
based
SQL
based
databases.
There
are
no
SQL
databases,
there
are
graph
databases,
okay,
so
every
application
is
going
to
work
around
data.
Okay.
So
also
even
my
recent
experience
at
Google
I've
been
working
on
these
kind
of
applications
and
one
problem
that
I
was
solving.
That
was
basically
that
we
had
huge
amounts
of
data
and
we
have
to
do
some
kind
of
a
string
searching
in
them.
H
Okay,
maybe
we
have
to
search
for
emails
of
maybe
10K
users
in
in
a
large-
let's
say
large
number
of
text
reports
Okay,
so
so,
generally
most
of
the
times,
you
would
find
very
common
applic
data
structures
like
list
array
or
map.
They
would
solve
your
purpose,
but
one
specifically
for
this
particular
application.
We
have
to
come
up
with
a
design
that
optimizes
the
queries
for
large
number
of
users.
Okay,
so
at
that
particular
point
of
time
we
have
to
create
a
specific
try
based
data
structure.
H
So
try,
if
you
know,
is
a
prefix
tree
kind
of
a
data
structure
and
we
have
to
optimize
this
problem
using
a
prefix
tree.
Okay,
so
generally
you
won't
encounter
hard
data
structure
based
problems
regularly,
but
sometimes,
while
working
on
a
system
design
or
solving
a
problem
using
a
system
design,
you
may
need
to
think
of
a
optimized
data
structure
that
can
reduce
the
load
on
the
server
that
can
reduce
the
processing
time.
That
can
reduce
the
complexity.
H
H
For
example,
you
want
to
build
a
system
to
process
the
videos
that
are
uploaded
by
the
users
on
the
YouTube
okay,
so
you
need
a
processing
pipeline
that
will
involve
the
use
of
data
section
like
queues,
maybe
multiple
people,
people
are
uploading
their
videos
and
each
video
is
going
to
take
certain
time
to
get
processed.
So
what
you're
going
to
do?
H
You
are
going
to
put
those
videos
inside
a
queue
and
then
someone
is
going
to
pick
those
videos
from
the
queue
it
will
go
to
the
processing
system
and
maybe,
though
videos
processed,
and
that
then
it
is
stored
in
the
database.
Okay,
so
you
need
certain
kind
of
data
structures.
Are
the
add
your
server,
which
can
persist
the
information
while
it
is
being
processed?
Okay,
so
in
every
system,
design
problem
is
going
to
involve
some
hidden
use
of
data
structures.
H
Even
if
you
are
making
a
game
like,
let's
say,
a
simple:
zero
cross
game:
okay,
maybe
tic-tac-toe.
So
what
data
sector
you
will
need?
Who
will
tell
me
what
data
sector
you
will
need
to
make
this
kind
of
a
game?
Okay,
so
simply
you
will
need
a
2d
array.
Okay,
2D
array
is
a
data
structure.
Okay,
2D
array
is
a
data
sector
or
maybe
you
are
working
with
images
in
machine
learning.
H
Let's
say:
let's
talk
about
machine
learning,
you
are,
you
are
going
to
process,
let's
a
lot
of
images,
so
each
image
is
nothing
but
it's
a
3D
Matrix
of
pixels.
Okay.
So
if
you
have
a
image
of
let's
say
one:
zero,
eight
zero
by
720
and
each
pixel
is
represented
using
three
values:
let's
say:
RG
and
B,
maybe
255
200
and
80..
So
you
actually
need
a
3D
array
data
structure
to
store
one
image
inside
the
ram.
Okay.
H
That
is
something
that
we
do
very
frequently,
and
sometimes
this
data
can
be
in
the
form
of
your
video.
So
a
video
is
nothing,
but
it's
a
list
of
images.
Okay,
so
if
it
is
a
list
of
images,
then
each
image
as
I
said,
is
a
3D
array.
Then
list
is
adding
one
more
Dimension,
so
this
data
can
be
a
4D
data,
okay,
so
a
sequence
of
frames.
So
we
have
a
4D
data.
The
fourth
dimension
is
the
dimension
of
time.
H
Okay,
so
we
will
need
a
four
dimensional
data
structure,
a
four
dimensional
array
to
store
this
data.
Okay,
yes,
so
red,
green
and
blue,
so
RGB
colors
is
it's
a
RGB
Matrix.
You
can
actually
see
in
the
code
itself
if
you
are
familiar
with
python
or
you're
working
with
a
library.
There
is
a
library
known
as
numpy.
There
is
a
library
opencv
you
can.
You
can
actually
read
any
image
and
see
the
underlying
array
behind
that
particular
image.
Okay,
so
data
structures
play
a
very
important
role
in
system
design
in
designing
applications.
H
Okay,
so
let's
talk
about
the
application
like
zomato,
then
maybe
I
I
tell
him
I
tell
the
app
actually
I
want
to
find
all
the
orders.
Maybe
I
want
to
find
all
the
items
that
are
offered
by
a
restaurant,
XYZ,
okay
or
let's
say
I-
want
to
find
out
all
the
pizzas
that
are
offered
by
Domino's
okay,
so
one
thing
is:
zomato
is
going
to
maintain
a
list
of
restaurants,
R1
R2,
R3,
R4,
R5
and
so
on.
H
Okay
and
each
restaurant
has
multiple
items
now
suppose
if
I
want
to
find
out
what
are
the
items
offered
by
this
particular
restaurant,
so
what
I
am
going
to
do?
What
kind
of
data
structure
might
be
useful?
Okay,
so,
instead
of
keeping
it
as
a
simple
array,
we
can
actually
maybe
use
a
hash
map.
We
can
say
Okay
I
want
to
find
out
what
is
offered
by
RX
this
restaurant.
So
we
can
put
all
these
values
as
items
that
are
offered
by
this
particular
restaurant.
Okay.
H
So
a
simple
data
structure
like
hashmap
could
be
useful
in
this
particular
case.
Okay,
so
knowing
what
data
structures,
what
operations
we
can
perform
on
different
data
structures
will
help
you
to
decide
your
application
logic
and
that
is
going
to
optimize
optimize
your
app
as
well
okay.
So
if
you
return,
if
you,
if
your
app
is
fast,
then
you
are
going
to
have
a
good
user
experience
as
well,
because
people
don't
want
to
wait
for
their
orders.
H
People
don't
want
the
page
to
be
loading
okay,
so
we
want
to
give
good
user
experience
and
the
back
end
for
building
a
good
user.
Experience
requires
extensive
use
of
data
structures.
Okay,
yes,
indexing
will
also
be
able
to
help
so
yeah.
So
maybe
you
want
to
build
a
application
like
uber,
then
storing
this
data
may
require
a
different
kind
of
a
data
sector.
Okay,
now
the
problem
is
not
about
finding.
Let's
say
we
have
lots
of
caps
on
Uber,
let's
just
from
C1
to
C1
million.
H
We
have
lots
of
cabs
and
maybe
a
user
is
standing
here.
Then
how
do
you
find
out?
What
are
the
nearest
camps
that
are
available
to
this
particular
user?
Again,
this
can
be
a
data
structure
problem.
You
will
need
to
find
find
out
what
all
caps
are
available,
what
cabs
are
available
in
this
region
and
what
tabs
are
available
to
this
particular
user?
So
again
we
have
to
think
and
come
up
with
an
algorithm
which
computes
it
in
a
very
fast
time.
H
Okay
or
maybe
you
want
to
find
a
route,
maybe
the
cab
is
a
shared
ride,
and
maybe
you
want
to
find
out
what
co-passengers
should
I
pick
so
that
the
time
of
the
trip
is
reduced?
Okay,
if
it
is
a
shared
right
from
what
locations
should
I
pick
the
co-passengers,
so
that
the
time
of
the
trip
is
reduced
again,
it
would
be
a
problem
that
can
be
solved
by
the
knowledge
of
competitive
programming
and
data
structures,
and
you
have
to
come
up
with
a
design
to
solve
this
kind
of
a
problem.
H
Okay,
so
there
might
not
be
a
straightforward
forward
algorithm.
There
might
not
be
the
best
algorithm,
but
maybe
you
can
come
up
with
the
approximate
algorithm
that
gives
you
close
to
the
best
results:
okay,
so
yeah!
So
that's
another
application:
okay,
even
in
machine
learning,
applications
like
computer
vision
and
all
okay.
So
there
is
a
concept
of
image
segmentation.
Okay.
H
Nowadays,
this
image
segmentation
is
done
using
deep,
deep
neural
networks
which
are
based
on
cnns.
But
earlier,
if
you
go
back
few
years,
the
image
segmentation
was
done
using
graph
algorithm,
so
I
will
give
you
one
example.
So
earlier
this
image
segmentation
may
be.
Let's
say
there
is
a
cat
in
the
picture
and
you
want
to
identify
what
pixels
of
this
picture
belong
to
the
cat
Okay.
So
segmentation
is
nothing
but
it's
a
pixel,
wise
labeling.
H
H
Has
anyone
heard
about
this
topic?
Network
flow,
Min,
Cut,
Max
cut
so
I
have
a
graphs
course
on
udemy
that
talks
about
this
topic,
and
this
topic
was
used
to
build
a
real
life
application
of
segmenting.
What
pixels
belong
to?
What
object?
Okay,
so
you
can
see
again
it's
a
topic
that
generally
people
don't
study
in
data
structures,
but
it
is
something
that
I
teach
in
competitive
programming:
okay,
so,
okay!
So
if
you
understand
graph
algorithms
in
depth,
you
might
be
that
might
help
you
in
future.
H
If
you
go
into
ml
side
of
things
or
machine,
deep
learning,
side
of
things,
and
although
nowadays
this
is
solved
using
deep
learning,
so
these
are
old
algorithms,
but
yeah
they
were.
They
are
still
used
in
certain
applications.
Okay,
so
the
problem
with
deep
learning
is
they
require
a
lot
of
data.
But
if
you
don't
have
a
large
data,
then
you
have
to
come
up
with
techniques
like
these
okay.
H
Okay,
so
now,
through
these
a
lot
of
examples,
what
we
have
observed
is
that
problems
involving
data
will
require
how
the
data
is
stored
and,
depending
on
how
the
data
is
stored,
we
can
we
perform
different
operations
on
the
data.
Okay,
so
create
we
create
new
data,
we
read
data,
we
update
the
old
data
and
we
also
delete
the
data
okay.
So
these
are
operations
are
known
as
crud
operations
and
every
application
will
involve
doing
these
crud
operations.
Some
some
applications
may
be
read
intensive.
Maybe
you
are
reading
a
lot
of
data
sub
intensives.
H
Some
applications
might
be
right
in
intensive.
That
means
you
are
writing
a
lot
of
data
okay.
So,
for
example,
let's
talk
about
YouTube
okay,
so
maybe
one
percent
of
the
people
are
creating
videos.
99
of
the
people
are
watching
videos,
so
that
means
a
lot
of
reads
are
happening
as
compared
to
the
number
of
Rights.
So
accordingly,
your
data
structure
should
be
optimized
to
support
these
99
of
the
people.
Okay,
we
don't
want
a
data
structure
which
optimizes
for
rights.
H
We
want
to
optimize
the
reads:
okay,
yes,
so
all
web
applications
will
involve
these
crowd
operations
as
well.
Okay,
so,
depending
upon
the
application,
the
data
may
be
in
the
form
of
an
array
it
may
be
in
the
form
of
a
hash
map.
It
may
be
in
the
form
of
a
priority
queue.
Let's
say
nowadays
we
know
that
covid-19
Drive
is
going
on
and
maybe
out
of
10
million
people.
We
want
to
prioritize
the
1
million
people.
H
We
want
the
software
to
come
up
with
the
1
million
people
who
are
maybe
in
the
highest
age
group,
and
we
want
to
give
them
the
shots
first,
okay,
so
in
that
case
an
application
like
priority
view
may
help,
because
it
solves
this
problem
in
an
optimized
time.
We
do
not
need
to
sort
the
entire
list
of
people.
We
can
build
a
heap
and
find
out
the
top
million
people
in
from
that
as
well.
Okay,.
H
Okay
and
we
talked
about
graphs,
so
a
graph
graphs
are
also
used
in
databases,
as
I
said.
One
example
of
graph
database
is
a
neo4j
and
applications
like
Facebook
Twitter.
They
need
these
kind
of
graph
data
structures
to
optimize
their
queries.
Maybe
I
want
to
find
friends
which
are
people
who
are
connected
with
me,
and
they
are
interested
in
watching
these
kinds
of
movies.
Okay,
so
you
will
not
run
a
query
on
a
billion
people.
You
will
run
only
you
will
only
query
the
people
who
are
maybe
connected
with
me.
H
Okay,
one
degree,
the
first
degree
connections
Okay.
So
this
is
one
another
area
where
these
data
structures
might
be
used.
Okay,
so
what
is
the
advice
I
would
like
to
give?
Give
it
to?
You
is:
do
understand
the
basics
of
data
structures
really
very
well.
They
will
help
you
in
becoming
a
good
software
developer
regarding
competitive
programming,
whether
you
want
to
go
into
the
advanced
data
structures
or
solve
problems
on
code
forces
or
not
I
would
totally
leave
it
to
you.
Okay.
H
So
if
you
are
interested,
if
you
are
passionate
about
solving
those
kinds
of
problems,
then
you
can
do
competitive
programming,
but
it
is
optional.
Okay,
but
overall
it.
It
is
a
good
brain
exercise.
It
is
a
rewarding
experience.
It
will
definitely
help
you
in
coding
interviews
it
would.
It
is
good
for
college
students
and
Learners
once
you
are
into
the
industry.
You
don't
need
to
spend
time
into
these
competitive
programming
problems,
because
you
would
have
the
own
problems
of
your
own
on
on
your
own.
H
Okay
and
taking
part
in
contests
gives
you
a
sense
of
achievement.
Okay,
people
write
in
the
resume
that
I
won
a
10th
rank
in
this
particular
contest,
and
even
companies
see
that
okay,
what
you
have
achieved
in
Google
Kickstart,
what
you
have
achieved
on
code
forces,
okay
and
then
it
also
helps
in
online
hiring
rounds
online
hiring
rounds.
If
you
see
a
lot
of
companies
are
asking
questions
on
competitive
programming.
H
So
if
you
want
to
practice
for
those
rounds,
then
definitely
do
some
coding
contests
and
learn
Advanced
topics
which
sometimes
may
or
may
not
be
useful.
So,
as
I
said,
it's
like
driving
a
car
in
the
F1
race
that
car
might
not
be
used
in
in
your
real
life.
So
similar
is
with
the
competitive
programming.
It
will
help
you
to
clear
the
online
rounds,
but
it
might
not
help
you
with
the
real
software
development
that
you
generally
do
so.
Software
development
is
more
real.
H
You
build
the
real
software
for
the
real
users
and
it
is
good
for
societies.
People
get
to
use,
people
can
use
a
lot
of
applications
and
you
can
solve
real
life
problems.
Okay,
and
there
is
also
a
steep
learning
curve.
Once
you
enter
this
software
industry,
there
are
lots
of
Frameworks
lots
of
new
technologies
that
are
coming
in
emerging
in
and
you
get
to
get
to
learn
a
lot
and
you
get
to
solve
new
and
challenging
problems
every
day
you
have
to
come
up
with
the
designs
of
new
systems.
H
As
you
progress
through
your
career
and
a
lot
of
things
you
can
do
in
software
development
as
well.
Okay,
so
for
college
I
would
recommend,
maybe
for
a
couple
of
months,
maybe
six
to
eight
months.
Definitely
do
data
structures
and
algorithms
really
very
well
if
you
are
interested
into
CP,
go
into
that
otherwise
some
amount
of
web
development
or
Android
or
some
other
stack.
H
You
should
also
do
at
least
for
three
to
six
months
to
get
a
hang
of
real
software
development,
and
this
is
something
that
is
going
to
help
you
to
build
projects
for
your
resume,
and
this
is
something
that
companies
nowadays
are
also
appreciating.
Okay
hiring
is
moving
from
competitive
towards
web
as
well.
Okay.
So,
ideally,
you
should
have
both
the
both
these
skills
in
your
resume.
Okay,
so
do
both
you
can
do
60
on
competitive
40
on
dev
or
you
can
do
40
on
CP
and
sixty
percent
of
div.
H
So
this
is
something
that
I
would
recommend.
Ideally,
50
50
is
recommended.
Okay,
and
you
should
manage
your
college
time.
Accordingly,
data
success,
algorithms,
third
year
method,
little
bit
of
web
development
or,
if
you're
starting
from
the
first
year,
then
you
get
an
edge
because
you
have
lot
of
time
to
master
these
skills.
H
Okay,
so
generally
12
to
16
months
is
good
enough
time
to
master
both
of
these
skills,
and
apart
from
it,
you
have
lot
of
other
semesters
as
well,
in
which
you
you
can
learn
about
new
topics
like
certain
Frameworks,
like
Android
or
machine
learning,
or
data
science
or
working
with
Version
Control.
So
a
lot
of
things
you
can
achieve
in
your
college
if
you
start
from
your
first
semester
itself.
Okay,
so
that's
all
about
my
talk
for
today,
I'm
getting
some
questions.
I
am
in
the
10th
10th
grade.
H
H
H
Okay,
so
all
these
courses
are
there
on
my
website,
codingminutes.com
and
all
of
these
courses
cost
less
than
a
Pisa.
You
can
pick
any
course
of
your
choice.
You
can
do
it
and
that's
also
going
to
help
you
so
I'm,
bringing
a
course
on
competitive
programming
for
those
who
are
interested.
Web
development
will
also
come,
but
yeah,
okay,.
H
So
people
are
asking
me
for
the
mail
ID,
so
my
mail
ID
is
pratik
at
the
rate.
Codingminutes.Com,
okay
and
you
can
also
connect
with
me
on
LinkedIn
and
I,
also
have
a
Gmail
ID,
which
is
pratik
narang,
triple
one
at
gmail.com,
okay,
so
that
that
was
about
the
topics
that
I
wanted
to
touch
today.
I
hope
you
gained
some
knowledge,
some
insights
about
competitive
programming
and
software
development.
So
in
real
you
have
to
do
both
once
you
graduate.
You
would
be
focused
on
software
development,
but
during
your
college
you
would
also.
C
C
C
C
C
C
C
B
So
I
think
we
have
the
five
winners
on
the
school
board
for
the
band
yeah
for
the
band
One
quiz.
So
there
was
a
quiz
before
this
particular
quiz,
so
you
can
check
the
I
think
the
school
good
for
it.
Okay,
so
whoever
the
winner
just
ping
us
on
our
mail
ID,
we
will
get
back
to
you
for
further
details,
but
we
need
and
how
to
process
and
I
think
we
are
good
to
go
so.