►
From YouTube: Rethinking Cloud Development
Description
MetalBear builds tools for backend developers. Their software is open-source, self-serve, and designed with one objective in mind - to make backend developers’ lives a little better with every release.
In this Open-Source Friday Stream, join Aviram Hassan as he demonstrates how MetalBear is transforming cloud development
Host: Andrea Griffiths
As always, feel free to leave us a comment below and don't forget to subscribe: http://bit.ly/subgithub
Thanks!
Connect with us.
Twitter: http://twitter.com/githubcommunity
Meeetup: https://www.meetup.com/github/
https://twitter.com/aviram
https://twitter.com/ALAColombia
About GitHub
GitHub is the best place to share code with friends, co-workers, classmates, and complete strangers. Millions of people use GitHub to build amazing things together. For more info, go to http://github.com
A
A
I'll
write
a
comment:
yeah
I'd,
read
a
comment.
Thank
you.
So
much
tiles
from
meadowbear
for
letting
us
know
perfect.
Well
we're
just
gonna,
rewind
and
restart
and
hi
everyone.
I
think
it's
friday
and
that's
that
no,
the
computer
doesn't
want
to
work
anymore.
The
mics
don't
want
to
work
anymore,
but
I'm
happy
that
you're
here
with
us
today,
because
we
have
a
treat
for
you
if
you're
interested
in
backhand
tools
or
cloud
engineering
you're
at
the
right
place,
we're
going
to
learn
from
avram
again
hi
welcome.
Thank
you
for
joining
us
from
metal.
B
A
B
A
Fantastic,
thank
you
so
much
and
thank
you
for
being
here
because
I
know
it's
kind
of
late
for
you
as
well.
This
is
I
go.
Listen
I
go
to
bed
at
eight,
so
that
is
late
for
me,
but
thank
you
for
being
here.
We're
super
excited
to
learn
about
your
project.
I'm
gonna
ask
a
few
questions
and
then
hopefully
our
audience
will
also
have
some
questions
to
ask.
A
B
Yes,
so
metal
bell
is
a
bacon
tools
company
and
we
founded
it
for
developing
open
source
tool
for
backend
engineers.
We
are
global
and
remote.
First
with
team
members
from
brazil,
canada,
germany
and
israel.
Actually,
our
first
team
members
were
recruited
from
reddit
based
on
their
open
source
contributions
on
github,
with
no
pro
or
professional
experience.
B
B
What
meredith
does
it
lets?
You
walk
on
a
remote
environment
from
your
local
setup
seamlessly
we
started
working
on
mrd
from
around
february
and
the
code
is
open
source
and
we
started
building
the
community.
We
even
started
doing
releases.
We
are
we're
currently
at
version
2.11
and
we'll
soon
to
release
our
next
major
version
version
3,
which
will
have
tons
of
new
features
and
hopefully
we'll
solve
many
problems
that
backend
engineers
face
in
their
daily
lives.
A
Awesome,
thank
you
so
later
on.
I
do
want
to
ask
you
like:
where
did
this
idea
come
from?
I
think
a
lot
of
things
and
I
love
that
the
story
about
your
how
your
hire
happened.
This
is
why
it's
so
important
folks
that,
if
you're
out
there
contributing
to
open
source,
you
know
that
this
actually
leads
to
opportunities
right.
I
love
that
you
actually
hire
someone
based
on
their
commits
and
what
they
were
doing
for
the
project,
so
we're
going
to
dig
into
all
of
that.
A
A
B
A
B
So
this
is
just
showing
the
setup,
so
we
set
up
a
kubernetes
cluster
with
a
demo
of
of
an
architecture
by
wevox,
it's
a
it
has
many
microservices
and
the
setup
is
like
the
user.
You
have
a
front-end
service
that
is
a
node.js,
and
then
you
have
tons
of
other
back-end
services
in
different
languages
and
with
different
logic,
and
for
this
demo
we're
just
going
to
focus
on
the
first
back-end
service,
the
back-end
service,
that
provides
the
front-end
for
ease
of
understanding.
B
Yeah,
so
we
have
all
the
services
running
and
everything
is
working,
so
we
can
just
go
on
to
the
site
and
browse,
and
this
is
of
course,
very
simple.
We
have
services
online,
we
can
browse
and
use
it,
and
the
idea
with
mercy
is
that
I
can
in
many
times
when
you
develop
software,
you
want
to
see
how
it
behaves
in
real
life,
and
for
that
you
either
run
it
locally.
And
then,
when
you
start
having
a
more
complex
architecture,
you
need
a
lot
of
requirements.
B
For
example,
if
you're
running
off
azure,
you
would
have
many
services
of
azure,
like
storage,
like
q
services,
databases,
and
then
it
becomes
sometimes
even
unmanageable
to
set
up
all
of
this
locally
all
for
each
developer
and
the
idea
is
to
create
a
solution
that
lets
the
developer
work
on
the
remote
environment
from
the
comfort
of
their
own
machine.
So
we
have.
B
And
what
you
can
see
do
actually
is
basically
the
most
basic
feature
is
to
run
the
process
with
remote
traffic.
So,
instead
of
starting
to
test
with
different
parameters
and
trying
to
mock
what
will
happen
in
a
real
environment,
you
can
just
run
the
service
locally,
see
how
it
behaves
with
real
traffic.
Coming
from
your
remote
environment
and-
and
the
next
features
we
had
in
community
and
some
are
already
available-
is
to
actually
have
more
of
the
context
of
the
cloud.
B
For
example,
you
would
have
access
to
databases
as
if
you're
running
on
the
remote
code,
you
would
have
access
to
files
to
environment
variables
and
I'll
show
this
or
all
in
a
bit.
So
we
have
here
the
code
of
the
frontend
service
and
we
have
mirrored
the
extension
here,
so
you
can
see,
let's
open
it,
so
you
can
see
here
the
disabled
energy
and
we
can
also
enable
it.
Let's
leave
it
enabled
right
now
and
now,
when
I
can,
I
can
click
on
debugging
the
service,
and
then
it
will
start
running
it
will.
B
So
I
want
to
impersonate
the
remote
front-end
pod
and
I
will
click
on
that
and
now
I
will
wait
for
it
to
start
running
and
we
see
here
that
it
started
running
and
I've
set
here
breakpoints
from
previously
and
we
can
see
that
requests
that
come
to
the
catalog
should
be
intercepted,
not
intercepted,
but
mirrored
so
it's
what
we
will
see
that
the
website
will
still
work,
but
I
will
be
able
to
inspect
the
request
and
see
how
it
happens
and
how
it
interacts
with
my
code.
So
let's
just
do
that.
B
B
It's
not
that
I'm
debugging
the
remote
instance,
but
I'm
mirroring
the
traffic
coming
to
the
real
service,
and
then
I
can
walk
locally
in
the
traffic
and
see
how
it
behaves
and
modify
it
test
it
a
bit
and
then,
when
I
feel
comfortable
enough,
I
will
be
able
also
I
I
can
be.
I
would
be
able
also
to
steal
the
traffic
and
then
see
how
it
really
behaves
in
the
remote
environment.
B
B
B
And
a
bit
about
more
elaborate
options,
so
you
can
do
my
mute
here,
so
we
have
already
tcp
still
we
have
override
and-
and
we
have
enabled
tcp
outgoing,
which
means
that
I
can
access
remote
services
on
behalf
of
the
board
and
remote
dns.
So
the
reason
it's
only
as
feature
flags,
because
it's
not
that
stable.
Yet
we
as
we
said
we're
working
on
it
a
lot
and
we're
also
working
in
central
semantic
versioning.
So
we
don't
want
to
introduce
breaking
changes.
So
we
just
add
more
features.
B
We
let
users
enable
them
of
themselves,
but
when
version
3
comes
in
then
most
of
the
features
will
be
enabled
by
default.
So
you
just
do
melody
exec
and
you
would
you
would
get
the
remote
dns
remote
traffic,
a
mirroring
and
the
environment
variable,
so
you
could
just
plug
into
a
remote
power
pod
that
is
already
running
without
needing
to
fetch
the
environment.
B
Variables
understand
what
databases
you
need
access
to,
what
files
you
need
to
have
locally
what
permissions
and
you
can
just
start
working
on
real
traffic
and
see
how
it
behaves
and
when
you
feel
more
comfortable
with
the
api
is
coming
in
and
api
is
going
on.
You
can
enable
the
tcp
still
and
actually
be
the
service
that
returns
the
response
to
the
actual
user
yeah.
So
I
hope
it
made
sense
any
questions
so
far.
A
A
You
know
there
is
a
ton
of
stuff
out
there
right
now
being
written
about
in
production
testing
and
how
like
now,
you
shouldn't,
spend
so
much
time,
creating
all
these
staging
testing
scenarios,
where
just
going
into
production
and
being
able
to
like
what
this
tool
does
right
being
able
to
mirror
and
then
test
and
not
have
the
site
down.
That's
that's
just
amazing,
quick
question,
though
so
when
you're,
when
you're
starting
this,
can
you
control
like
what
what
you?
What
cluster?
Actually
you
access
like
what
you
impersonate?
A
B
Okay,
so
let's
do
a
bit
of
a
technical
dividend
because
it
might
be
interesting
to
the
to
you
and
the
audience.
Yes,
so
the
way
melody
works.
It
has
a
about
two
main
components:
let's
call
it
so.
The
the
the
first
component
is
the
component
that's
running
on
the
machine
of
the
developer
and
that
component
executes
the
process
and
also
loads
to
the
process.
B
That
decides
like
okay,
I
want
to
remote
traffic.
I
want
local
traffic,
I
want
remote
files,
local
files,
and
I
can
actually
control
okay.
If,
if
it
has
the
side
effects,
I
can
disallow
it
or
allow
it,
and
on
the
cluster
side,
we
have
an
agent
running
that
is
spawn
automatically
using
the
cli
and
you
don't
need
to
install
and
pre-install
anything.
Apart
from
the
cli
on
your
desktop
or
the
extension
and
the
cli
spawns
the
agent
and
the
agent
access
proxy.
So
whatever
they,
the
local
local
machine,
wants
to
do
with
the
local
process.
B
The
the
agent
on
the
cluster
side
is
responsible
for
actually
executing
and
performing
this,
and
the
agent
runs
in
the
context
of
the
remote
pod.
So
it's
as,
if
you're
running
on
the
pod
itself,
so
it's
basically
having
the
closest
you
can
to
re
to
run
in
real
life
in
production
without
deploying
that.
A
All
right,
that's
awesome.
Thank
you
so
much
that
actually
paints
a
picture
that
I
can
see
and
understand.
Thank
you.
I
appreciate
it
and
and
folks
who
are
watching.
If
you
have
any
questions,
I'm
sure
avidan
wouldn't
mind
digging
into
more
of
the
technical
aspects
of
how
he
built
this
thing.
So
don't
be
shy,
go
ahead
and
ask
any
questions.
A
I
also
share
a
blog
post
that
was
on
your
website
from
when
you
were
introduced,
mirror
d
back
in
may,
which
has
got
some
excellent
graphics,
making
the
case
for
testing
and
production,
which
is
really
great.
So
I
do.
I
love
the
dynamic
there
because
and
the
tone
of
the
post
is
amazing,
so
go
ahead
and
give
it
a
read
folks.
Why
did
you
make
this?
I
mean.
Usually
these
things
happen
because
you
have
a
problem
that
nothing
out
there
is
solving,
but
I
love
to
hear
like
what
was
your
motivation
behind
building
this.
B
Yeah
so
me
and
ayal,
my
co-founder
and
partner
at
metal
belt.
We
worked
together
at
a
fintech
company
called
biocatching,
and
we
were
at
the
back
in
group
and
we
experienced
a
lot
of
issues
that
plagued
the
daily
work
of
back-end
developers.
We
were
really
annoyed
by
that
and
we
fought
like
we
underserved
like.
We
saw
many
tools
for
devops
and
many
tools
for
other
departments
and
for
engineers,
there's
like
the
ide,
debuggers,
etc,
but,
like
we
didn't,
have
like
any
tool
that
helped
us
work
on
our
it.
B
We
wanted
to
create
a
tool
that
lets
you
interact
with
staging
and
production
from
day
one
instead
of
meeting
it
after
all,
the
ci
cd
steps,
which
are,
of
course
necessary,
but
sometimes
the
cicd
just
slows
you
down
from
seeing
how
it
behaves
in
real
life
and
in
real
life.
All
the
code
you've
done
all
the
tests.
B
You've
done
doesn't
really
matter
because
you
mismatched
an
sdk
version
and
the
cloud
service
you're
using
is
different
and
there's
a
lot
of
increases
in
consistency
there
and
we
want
also
to
solve
the
problem
of
developer
environment.
So
right
now
you
set
up
in
big
companies.
You
set
up
a
developer
environment
developer
and
I'm
not
talking
about
the
developer
space
like
code
space,
because
it
doesn't
really
we.
B
We
can
work
with
code
space,
because
cloud
space
allows
you
to
have
some
ide
to
work
on,
but
you
won't
run
the
old
cluster
on
the
on
the
code
space
because
you
would
need
managed
services,
you
need
state
and
that's
the
real
problem
right.
It's
not
to
set
up
the
environment
in
in
big
companies.
You
have
infrastructure,
schools
or
setting
up
environments
is
easy,
but
working
and
doing
it
at
scale.
That's
the
problem
and
that's
like
our
vision
for
it.
B
A
B
So
actually,
we
have
a
few
great
contributors
that
are
also
doing
it
on
a
weekly
daily
basis
and
we're
accepting
more,
of
course,
and
just
to
make
it
clear.
I
believe
that
open
contributing
to
open
source
is
not
just
writing
code,
like
we
had
people
fixing
the
documentation
and
getting
started
at
the
website.
B
B
This
is
not
well
explained,
so
feel
free
to
open
issues,
and
if
you
want
also
to
fix
those
and
suggest
changes,
that's
that's
more
than
welcome
and
we
appreciate
that
and
we
have
the
getting
started
instead
get
installed
for
contributions
in
case
you
want
to
get
into
code
and
we
are
also
willing
to
mentor
new
people.
We
have
people
that
started
learning
rust
using
our
code
base
and
we
they
joined
our
disco
channel
and
then
we
started
talking.
They
said
hey.
B
We
want
peace
ticket
and
we
helped
them
do
the
sales
issue
on
github
and
provided
them
with
guidance.
And
that's
that's
great.
I
think
it's
good
it's
great
from
our
side,
because
then
we
can
see
the
blind
spots
like
real
experience.
We
know
the
project
and
having
someone
coming
from
the
outside
and
getting
on
getting
started,
something
that
we
do
have
because
we
recruited
new
engineers
on
some
basis
but
like
having
people
regularly
doing
the
onboarding
and
starting
to
work
really
gives
us
the
overview
of
what
is
lacking.
B
A
B
But
we
want
to
to
be
the
productivity
company
for
backend
engineers,
so
we
see
hashicorp
as
like
a
role
model
like
hashicorps
for
devops
and
nashikops
developed,
so
many
tools
for
devops
that
develops
love
like
if,
if
there
was
a
tool
developed
by
hashicorp,
they
would
go
for
that
and
not
follow
the
competitor,
and
we
want
to
do
the
same
for
back
engineers.
Of
course,
we
start
small,
not
so
small,
but
with
the
first
project.
A
If
you
want
to
contribute
with
just
using
the
product
onboarding
on
it
and
giving
feedback
a
fresh
set
of
eyes,
that's
amazing,
then
you
can
go
to
the
repo
I've
put
the
rip
on
the
chat
notes
and
go
ahead
and
check
out
their
contribution
guidelines
and
get
in
there,
and
I
appreciate
you
mentioning
that
you're
welcoming
also
other
contributions
other
than
code,
that's
great!
So
if
folks
are
into
documentation-
or
even
you
know,
even
the
organizing
of
a
discord
server
and
like
running
that
community,
that's
also
a
way
of
helping
open
source.
A
A
B
B
The
world
map
is
to
start
by
elevating
developers
in
the
current
developer
environment
and
making
it
easier
for
them
to
work
on
the
remote
environment
and
then
move
to
share
the
environment
in
staging,
and
once
the
company
feels
safe
enough,
they
would
go
and
even
do
it
in
production,
and
once
we
gather,
we
will
already
have
enough
features
to
protect
you
and
also
role
management
and
access
management.
To
to
let
you
be
able
to
understand
production
and
also
on
the
strong
environment
that
is
not
production
without
causing
chaos.
A
Awesome,
thank
you
so
much.
I
appreciate
the
answer
and
thank
you
timelapse
for
the
question.
Folks,
if
you
have
any
other
questions,
please
feel
free
to
drop
them
in
the
chat.
I've
also
shared.
This
is
the
direct
invite
to
that
discord
server.
So
I
hope
that's
okay,
avram.
I
should
have
asked
you
first,
no.
B
A
B
Just
starting,
if
you're
senior
or
if
you're,
I
don't
know
thinking
about
transitioning
to
back
end,
I
think
I
I
grew
up
in
online
communities.
So
I
really
like
this
kind
of
stuff.
I
learned
everything
from
the
internet,
so
if
it
could
be
a
of
help
to
anyone
also
feel
free
to
reach
out
on
vm
on
discord.
Linkedin,
may
I
don't
know
whatever
channel
you
prefer
smoke.
A
I'll
be
sure
to
also
go
ahead
and
drop
in
your
twitter
handle
so
folks
go
ahead
and
follow
aviram
follow
metal
bear
so
that
you
can
stay
tuned
on
all
the
new
releases
for
mirror
b
and
all
the
amazing
things
you're
working
on.
So
I
have
a
question
for
you.
So
if
I
want
to
join
your
community,
obviously
I'm
not
backing
is
not
my
specialty,
it's
not,
but
so
you're
open
to
people
that
are
just
trying
to
get
started
right,
like
that's,
that's
yeah
awesome!
I
love
that.
Thank
you.
A
So
much
and
folks
take
advantage
of
this
opportunity.
You
know
there
is
not
a
lot
of
open
source
projects
that
you
get
to
be
there
from
the
beginning
and
actually
be
a
part
of
that
growth.
So
this
is.
This
is
a
really
unique
opportunity
for
all
of
you
to
participate
in
the
project.
So,
if
you're
back
and
curious,
even
if
that's
not
your
thing
go
ahead
and
join
the
community
because
you'll
be
able
to
learn
awesome
all
right,
so
I
think
we
have
about
all
the
questions
from
the
audience.
A
I'll
keep
an
eye
on
the
chat.
So
if
there
is
any
other
question,
we
can
go
ahead
and
answer
it
for
those
of
you
who
might
have
joined
later
on
before
we
were
having
some
sound
issues
at
the
beginning,
but
this
is
recording
and
I
will
upload
the
recording
to
youtube
so
we'll
be
able
to
you'll
be
able
to
go
back
and
see
it
and
see
the
demo
on
your
own
time.
A
B
B
A
A
B
B
Yeah
cool
so
excellent
question,
so
pizza
for
sure,
yeah,
so
yeah
we
had
about
telepresence,
it's
we're
trying
to
solve
the
same
use
cases,
so
it's
actually
like
a
competing
product,
but
the
difference.
The
differences
are
that
a
we
run
on
the
post
level
like
telepresence
change,
I'll
tells
the
operating
system
all
runs
via
container,
so
it
needs
to
add
a
little
stuff
instead
of
gpn
and
we
work
just
on
the
process
level.
So
we
believe
it
provides
better
user
experience.
B
So
you
don't
have
to
modify
your
whole
system
and
you
feel
like
plugged
in
into
the
pod
with
the
file
access,
no
mounts,
no
bringing
environment
variables
manually.
So
that's
like
the
the
first
big
difference
and
second,
our
approach
is
that
we
want
to
be
as
non-intrusive
as
possible.
So
you
can
just
start
using
you
duplicate
the
traffic.
You
don't
intercept.
B
We
will
have
a
feature
to
the
to
yourself
that
you
can
still
traffic,
but
that's
like
our
agenda
is
to
be
something
that
can
be
ran
on
production,
so
without
even
interfering
and
developing
features
that
that
protect
you
from
from
interfering
with
each
other.
So
yeah,
that's
like
the
basic
we
can
go
and
elaborate
more,
but
that's
like
the
high
level
and
feel
free
to
reach
out
on
discord
or
any
github
discussion.
If
you
have
any
questions
about
it,.
A
A
Are
your
dm's
open
his
dms
are
over
there
you
go
so
that
I'm
sure
you'd
be
happy
to
like
answer
any
questions
and,
and
you
know,
get
excited
about
contributing
if
you're
into
back-end
development.
So
the
other
question
that
result
result.
These
are
tough
questions.
It
would
be
for
me
to
have
to
answer
this.
What's
your
favorite
beyonce
song,
or
do
you
have
a
favorite
beyonce
song?
Who
knows
you
might
not
be
a
beyonce
fan,
which
I
wouldn't
say
that
out
loud
here.
A
That's
a
classic.
I
agree,
yeah.
I
agree
all
right
well.
Thank
you
so
much
for
joining
us
today.
Thank
you
for
the
demo
folks.
I
totally
apologize
for
the
issues
at
the
beginning,
we'll
figure
out
what
went
wrong.
I
think
it's
just
friday
and
we're
just
ready
to
be
done
for
the
day.
I
I
do
want
to
know
what
your
favorite
beyonce
song
is
so
go
ahead
and
post
it
on
the
chat
I'll.
Do
a
shout
out
too.
Thank
you
very
much.
A
Everyone
who
joined
everyone
who's
watching
the
recording,
don't
forget
to
follow
avidam
on
twitter
and
go
ahead
and
check
out
the
metal.
Bear
repository,
take
a
look
amir
d
and
the
contributory
guidelines
and
get
it
get
started.
This
is
an
exciting
thing
to
be
a
part
of
something.
That's
you
know
numerous
right
numerous.
So
that's
awesome!
Thank
you
so
much
many
closing
thoughts.
What
are
you
doing
this
weekend?
I'm
very
noisy,
I'm
way
nosier
than
roselle.
B
A
A
Thank
you.
Thank
you.
Everyone
who
watched
us
today
don't
forget
to
go
ahead
and
give
it
a
follow.
Go
to
the
project
on
github.
Give
it
a
start.
If
you
have
any
questions
post
them
on
the
chat
or
go
into
the
discussion
on
those
repos.
Thank
you
for
joining
us
for
open
source
friday.
It's
an
honor
to
be
here
with
you
now
go
enjoy
the
weekend.
I
said
you
can
be
off
now.