►
Description
In this session, Kevin Alwell, Principal Solutions Engineer at GitHub, and Jessi Moths, Director of Field Services Enterprise Products at GitHub, will talk about how you can increase developer happiness and productivity with GitHub Enterprise. They'll demonstrate how building a next-generation developer workflow can enhance and streamline the development experience using tools like GitHub Codespaces, GitHub Actions, GitHub Advanced Security, and GitHub Copilot effectively.
Stay up-to-date on all things GitHub by subscribing and following us at:
YouTube: http://bit.ly/subgithub
Blog: https://github.blog
Twitter: https://twitter.com/github
LinkedIn: https://linkedin.com/company/github
Instagram: https://www.instagram.com/github
TikTok: https://www.tiktok.com/@github
Facebook: http://fb.com/github
About GitHub
It’s where over 100 million developers create, share, and ship the best code possible. It’s a place for anyone, from anywhere, to build anything—it’s where the world builds software. https://github.com
A
A
A
We've
got
some
folks
moderating
to
help
with
that.
We've
also
got
a
button
up
there
for
our
requested
demo
or
schedule
a
meeting
with
us
if
something
here,
piques
your
interest
and
you'd
like
to
chat
more
with
someone
at
GitHub
we'd
love
to
do
that.
Go
ahead
and
use
that
button
to
do
so
as
far
as
viewing
goes.
A
Hopefully,
if
you're
seeing
this
and
hearing
this
you're
already
having
no
issues,
but
we
do
recommend
Chrome,
try
refreshing
if
you're
having
issues
with
the
stream
at
all
disabling,
the
VPN
can
also
really
help
or
an
incognito
window,
and
if
you
have
any
additional
issues,
there's
a
docs
tab
in
goldcast
that
you
can
use
we'd
be
happy
if
you
feel
comfortable
and
would
really
like
to
to
have
you
answer
or
ask
your
questions
in
video
q
a
later
on.
Here's
the
steps
on
how
to
do
that.
A
A
We
want
to
tell
you
what
we've
heard
from
customers
and
we
want
to
show
you
what
the
product
looks
like
looks
like
not
going
to
kill
you
with
slides
here
and
then
definitely
at
the
end,
like
we
said,
go
through
the
Q.
A
please
feel
free
to
put
your
questions
in
the
chat,
go
to
the
Q,
a
tab
up
at
the
top
there
to
make
sure
that
those
get
seen
so
that
we
can
answer
them
at
the
end,
all
right.
A
So
what
are
we
going
to
talk
about
today?
I
have
four
kind
of
General
big
areas
that
we're
going
to
talk
about
that
help
with
velocity
and
productivity
for
people
using
GitHub.
The
first
one
that
we're
going
to
touch
on
is
GitHub
issues,
but
also
how
they
integrate
with
our
built-in
project
management.
A
That's
something
that
a
lot
of
people
actually
know
about,
but
we've
got
a
really
flexible
but
powerful
work,
planning
and
tracking
system
right
in
GitHub
that
you
can
use
anytime
that
you're
using
GitHub
to
do
the
rest
of
your
work,
and
it's
really
cool.
So
I
wanted
to
take
a
chance
to
to
give
you
just
a
peek
at
that.
We're
of
course,
going
to
highlight
GitHub
actions,
which
is
our
CI
CD
and
workflow
automation.
Platform.
A
No
extra
infrastructure
overhead
required
to
use
that
Kevin's,
going
to
show
a
great
example
of
that,
as
well
as
GitHub
code
spaces,
which
are
cloud-hosted
full
development
environments
nearly
instantly
on
any
time
that
you
want.
Even
if
your
boot
strapping
a
really
big
project
and
you
can
pre-configure
them.
So
there's
no
more.
That
works
on
my
machine
kind
of
issue
that
you
get
sometimes
with
developer
environments
and,
of
course,
we're
going
to
talk
about
co-pilot.
Everybody
wants
to
talk
about
copilot
right
now.
A
It's
your
AI
pair
programmer
for
GitHub
and
helps
provide
contextual
coding,
suggestions
right
in
your
editor
as
you
work.
So
why
are
these
important?
Well,
some
of
these
help
your
devs
get
up
and
running
faster
right
so
for
the
example
from
Duolingo
which
you
may
have
seen
earlier
on
today
in
the
session,
they
increase
their
developer,
speed
by
25
using
codes
or
co-pilot,
rather
helping
to
limit
some
of
that
context.
A
A
So
Pinterest
saw
an
87.5
degree
percent
decrease
build
time,
moving
to
cicd
with
GitHub
actions.
Similarly,
with
LinkedIn,
their
developers
started
rating
their
processes
for
CI
CD,
with
a
40
increase
in
that
NPS
score
moving
to
using
GitHub
for
those
processes.
Blue
Yonder
has
a
30
minute,
deploy
window
down
from
an
entire
day
without
automation
of
CI
CD
and
at
CH
Robinson
they
were
able
to
move
to
162
percent
more
deployments
after
adopting
GitHub.
So
these
are
some
real
numbers
right
things
that
you
would
love
to
be
able
to
go.
A
Tell
your
leadership!
Hey!
We
have
this
return
on
investment
I'm,
not
making
any
of
this
up.
This
is
what
these
customers
have
really
told
us
that
they've
been
able
to
achieve
with
githubs,
so
we'd
love
to
be
able
to
talk
to
you
about
how
to
achieve
that.
For
your
organization
as
well,
and
then,
lastly,
you
know
thinking
about
not
only
the
internal
piece
making
internal
folks
happier
and
more
productive,
but
how
do
we
actually
ship
some
of
that
to
our
customers?
A
We
have
Walmart
that
are
shipping
up
to
14
000
times
in
a
month
with
GitHub
3M
went
from
400
deployments
over
60
days
to
15
000
and
their
developers
became
two
times
more
productive
when
they
moved
everything
to
GitHub.
Intel
saw
a
216
pull
request
growth
year
over
year,
so
more
than
doubling
the
number
of
times
that
people
collaborated
on
changes
and
Nationwide
saw
a
90
reduction
in
merge
time.
A
So
what
do
they
actually
say
about
these
different
things
for
issues
and
projects?
We
have
Shopify,
that's
also
using
this
quite
heavily,
and
they
really
like
that.
They
get
this
high
level
overview
across
all
their
projects,
able
to
triage
priorities
quickly
and
it's
familiar
to
their
Developers
code.
Spaces
helps
KPMG
to
be
able
to
deal
with
lots
of
different
requirements.
They
don't
have
to
do
all
of
the
tedious
error-prone
stuff
and
they
can
just
get
right
into
actually
doing
real
work.
A
Github
actions
at
Telus
has
saved
their
developers
around
two
hours
per
week
and
at
scale
that
multiplies
to
a
large
number
of
savings,
both
in
hours
and
in
dollars
at
Skyscanner.
They
report
that
their
their
Engineers
are
seeing.
You
know
ability
to
actually
work
on
business.
Logic
like
I,
said,
spend
time
on
the
questions
that
matter
most
that
actually
deliver
value
to
the
customer.
A
These
are
all
the
advantages
that
we
really
see
again,
moving
that
to
Velocity
that
velocity
needle
forward
and
keeping
the
happiness
while
you're
at
it,
not
just
driving
people
to
work
more,
but
doing
it
in
an
effective
way
that
allows
them
to
work
on
the
greatest
challenges
and
interesting
problems
that
they
want
to
solve
as
Developers
all
right.
So
we're
gonna
switch
over
here
now
to
talk
a
little
bit
about
a
day
in
the
life
of
somebody
working
on
GitHub
right.
A
What
does
this
actually
look
like
I'm,
going
to
play
the
role
today
of
of
a
product
manager
right
somebody
who's,
maybe
not
in
coding
every
day
on
GitHub,
but
needs
to
know
what
their
developers
that
they
work
with
are
doing
and
maybe
wants
to
manage
some
work,
while
they're
at
it
right
and
so
I'm.
In
my
applications,
repository
here
and
I'm,
starting
out
my
day
and
I
say
all
right.
A
I've
got
Sprint
planning
later
on,
let's
check
on
how
our
work
is
going
and
if
there's
anything
new
that
needs
to
be
assigned
to
anybody
to
work
on
so
I'm
going
to
come
over
to
my
projects.
Tab
here,
I'm
going
to
jump
into
my
tracking
board
that
we've
got
for
my
arapr
integration
application.
It
looks
like
we've
got
a
couple
of
items
here
that
don't
have
anybody
assigned
to
them
yet
to
work
on.
It
looks
like
my
engineering
manager.
A
Kevin,
probably,
is
the
one
to
work
on
these,
so
let's
get
him
assigned
and
do
that
very
quickly
here.
Just
from
the
same
view,
right
and
so
I
could
go
down
the
list
and
I
could
do
assignments
for
anybody.
I
needed
to
write
in
the
same
view,
let's
say
that
we're
going
to
start
out
our
our
Sprint
here
by
doing
a
little
bit
of
tea,
you
can
do
the
same
thing.
We
can
even
create
a
new
field
or
view
right
from
the
same
board.
A
If
we
wanted
to
now,
we
might
move
into
thinking
about.
Okay,
I
want
to
make
sure
that
my
work
is
balanced
across
my
work,
streams
or
I
want
to
see
where
the
different
velocities
are
for
my
different
work
streams.
I've
got
three
of
them
here.
I've
got
an
automation,
work
stream,
an
app
framework,
work
stream
and
a
delivery
work
stream.
It
looks
like
right
now.
The
bulk
of
the
work
is
being
done
in
that
app
framework
work
stream.
A
So
maybe
you
know
I
wanted
to
think
about
providing
them
some
extra
developers
to
share
time
something
like
that,
because
they've
got
a
lot
of
work
going
on
and
if
we
move
over
to
our
roadmap
View,
we
can
see
that
a
lot
of
that
work
again
is
being
centered
during
the
current
Sprint
that
we're
in.
So
we
have
both
kind
of
categorical
views
that
we
can
look
at.
We
can
actually
take
action
on
some
of
those
views
and
we
can
look
at
what
the
work
looks
like
over
time.
A
So
we
can
see
here
really
exciting
stuff.
If
we
jump
into
one
of
these
epic
items,
we
can
even
configure
subtasks
that
are
dependent
on
top
level
items.
So
we've
got
an
epic
here
with
subtasks
that
we
want
to
check
off
as
they
are
completed
all
of
these
roll
up
to
completion
of
this
single
epic-
and
this
is
all
done
right
in
GitHub,
with
the
issues
and
pull
requests
that
your
developers
are
used
to
using
it's
really
exciting
stuff
and
I'm
going
to
switch
over
now.
B
Right
on,
thank
you
very
much.
Jesse
hello.
Everyone
I've
had
a
lot
of
fun
already
interacting
with
you
in
the
chat.
It's
been,
it's
been
great
to
see
that
we
have
so
many
developers
present
from
all
over
the
world
and
I
think
there's
actually
been
some
specific
requests
already
and
so
feel
free
to
interact
with
us
directly
in
the
chat
and
I'm
fortunate
to
be
joined
by
some
of
the
great
folks
who
are
moderating
that
chat.
So
thank
you
for
doing
that
as
well.
One
of
the
things
that
I'd
like
to
address
that
came
up.
B
B
Think
a
lot
of
folks
have
been
curious
about
the
Synergy
between
chat,
GPT
and
GitHub
co-pilot
when
they
should
be
used
together
when
they
should
be
separate
and
what
the
best
use
cases
are,
and
so
copilot
chat.
I
think
is
a
nice
way
that
that
vision
for
a
large
language
model
integration
into
the
developer,
workflow
manifests
okay,
and
so
that's
somewhat
of
a
side
tangent
and
for
all
those
dark
mode
lovers.
I'm
bringing
us
back.
Take
off
your
sunglasses
Jesse.
B
There
was
a
lot
of
a
lot
of
spicy
opinions
about
dark
mode
or
not
in
the
chat.
So
it's
been
a
lot
of
fun.
So
let
me
talk
you
through
the
use
case
that
we're
going
to
demo
together
and
I.
Think
it's
a
pretty
interesting
one.
We
built
it
specifically
for
a
Galaxy
and
specifically
for
this
session.
So
hopefully
you
enjoy
the
conversation
and
learn
something
along
the
way.
B
Obviously
that's
the
goal
is
to
have
you
walk
away
with
a
little
bit
more
knowledge
and
you
came
into
the
session
with,
and
so,
if
that's
achieved,
then
we're
happy,
and
hopefully
you
are
too
so.
Basically,
what
we've
built
here
is
pretty
simple:
we
built
a
get
of
action
that
listens
for
when
pull
requests
are
opened
against
the
main
branch
and
it
synthesizes
the
change
that's
been
proposed
in
the
pr
and
it
contextualizes
the
pr
with
that
change.
B
Okay,
and
so
it
says,
hey,
this
pull
request
is
about
X
and
here's
the
effect
of
the
downstream
application,
and
so
basically
you'll
be
able
to
roll
this
action.
That's
built
in-house
for
your
organization,
Acro,
you
know
roll
it
out
across
your
entire
org
for
every
repository,
okay,
and
so
this
demo
has
a
few
layers.
B
The
first
thing
I'd
like
to
do
with
you
is
just
to
see
this
thing
live
once
so
it's
less
opaque,
more
tangible,
so
we're
actually
going
to
see
the
demo
live
once
then
we're
going
to
go
into
more
production
like
experience.
So
actually
you
know
in
development
mode
as
a
developer,
building
an
application.
So
it's
going
to
get
a
little
bit
more
complex
and
then
what
I
actually
wanted
to
show
you
is
kind
of
the
third
layer
of
the
demo.
Here
is
what
the
developer
experience
was
for
the
developer.
B
Like
yourself,
Ryan
out
there
I
saw
you
in
the
chat
and
Shaker,
and
so
many
others
we're
actually
building
the
automation
that
serves
developers
across
your
organization
right
and
so
I,
wouldn't
Focus.
So
much
centrally
on
what
this
integration
does.
Specifically,
it's
an
interesting
use
case,
but
it's
also
like
okay,
well,
what
tools
are
being
used
along
the
way
and
how
and
how
can
that
possibly
impact
my
own
developer,
workflows,
okay,
so
that's
the
whole
preface
that
I'll
give
you.
B
B
It's
really
just
a
test.
Repo
I
was
using
in
order
to
to
prepare
for
this
conversation.
This
event
that
we're
having
now
and
the
first
thing
that
you'll
notice
is,
we
have
our
workflows
directory
directory
nested
inside
dot,
GitHub
right,
and
so,
if
I,
just
poke
in
there
for
a
moment
and
show
you
what
the
get
of
actions
workflow
looks
like
we'll
take
a
moment
to
do
that
together
and
hopefully
the
site
of
code
isn't
intimidating
to
you.
B
I
saw
a
lot
of
devs
here,
so
I
feel
like
it's
not
and
that's
great,
because
we're
going
to
spend
some
time
in
it
together.
So
yeah,
the
name
of
of
your
workflow
here
right
is
AI
test.
It's
triggered
when
we
open
a
pull
request
against
our
main
branch,
and
it
runs
on
the
GitHub,
hosted
Linux
machines,
and
so
GitHub
actions
does
provide
compute
on
the
geta
platform.
So
you
don't
have
to
actually
run
those
agents
locally
or
as
part
of
some
gigantic.
B
So
let's
take
a
moment
together
and
actually
run
this
and
see
what
it
does
and
then
we'll
just
continue
to
layer
on
the
complexity
and
again
go
back
to
what
it
took
to
actually
build
this
and
we're
going
to
use
AI
to
build
AI.
It's
pretty
meta,
but
hopefully
it
keeps
things
interesting
and
fun,
and
so
maybe
your
experience
here
might
be.
You
know
I,
don't
know
you're
going
into
a
repository
and
you
want
to
scaffold
out
a
workflow
to
do
some
code
scanning,
so
I
might
click
into
my
security.
Tab.
B
I
might
look
to
enable
code
scanning
and
I
might
go
to
add
a
tool,
and
so
this
interface
is
actually
really
nice.
It
kind
of
flipped
us
back
to
the
actions
tab,
but
I
have
all
these
workflow
templates
available
to
me
now.
These
workflow
templates
are
published
by
folks
like
sneak
and
different
service
providers.
In
this
case
we
have
GitHub
for
code
ql,
but
you
could
also
publish
your
own
workflows
for
consumption
across
your
organization
to
package
up
those
modules
that
are
GitHub
actions
for
consumption
by
devs
or
as
a
developer.
B
You
have
some
idea
how
to
do
something
useful
with
the
new
language
models
make
that
available
to
your
org
right,
that's
kind
of
the
truest
form
of
inner
sourcing.
So
again,
let's
go
ahead
and
scaffold
out
this
code
tool.
Analysis
workflow,
so
we
go
ahead
and
hit
configure
here
and
I'm
going
to
just
commit
this
change
back
to
my
repository
I
might
make
the
commit
message,
something
like
I.
B
Don't
know
introduce
there,
we
go
introduce
code
scanning
via
code
ql
and
it
might
say
something
like
Leverage
codecule
to
scan,
rather
than
coming
to
the
main
branch
I'm
going
to
commit
to
a
branch
for
us
together
now
we'll
call
it
together
now,
Galaxy
together
now
I'll
go
ahead
and
propose
the
change
open,
the
pr
and
so
what
we.
So,
what
ought
to
happen
is
the
actions
that
are
listening
for
the
pull
request.
Event
should
begin
to
trigger
and
they'll
run
on
the
get
of
action.
So
there
we
go.
B
We
see
that
now
this
this
workflow
has
been
queued,
I'll,
go
ahead
and
click
into
details.
What
we'll
see
is
the
all
the
jobs
are
getting
set
up
in
this
case.
It's
just
a
single
job
which
is
test
Ai
and
we're
going
to
run
through
it,
and
so
you
have
your
live
logs
available
to
you
here
and
it
looks
like
it
actually
may
have
already
completed
so
live
logs
are
doing
some
catch
up,
so
I'm
going
to
go
back
to
my
pull
request
and
see
whether
or
not
this
was
successful.
Okay,
great!
B
B
So
I'm
not
going
to
read
that
to
you,
but
it
does
something
like
just
provide
additional
context
on
what
SAS
is
the
types
of
vulnerabilities
that
you
can
have
and
ensuring
that
you
know
your
application
meets
the
security
standard
set
by
the
organization
and
that
was
generated
on
the
Fly
I
had
no
idea
what
you
know
what
was
going
to
be
output,
I,
think
similar
to
what
you've
probably
already
experienced,
with
other
large
language
models
that
you've
tested.
Maybe
through
the
chat,
GPT
interface,
you
know
for
from
open
AI.
B
So
maybe
the
first
thing
that
I'll
do
is
see
if
any
environments
have
been
tagged
back
to
the
repo
and
have
been
actively
deployed,
and
so
obviously
that's
the
case
here
and
I
can
see
my
deployment
history
if
I
want
to
actually
view
my
deployment,
this
has
been
deployed
to
Azure
and
believe
it
or
not.
There's
the
bookstore
repo
represents
an
application,
that's
a
bookstore
which
is
wonderful,
and
so,
if
we
want
to
look
through
our
bookstore,
we
can
do
that.
B
We
see
different
titles
and
Status
Etc,
so
we're
actually
going
to
build
a
very
specific
feature
together,
which
is
the
rating
feature.
So
if
you
you
know,
if
you're
on
a
bookstore,
you
probably
want
to
see
what
people
think
about
the
different
books.
It's
a
good
Testament
to
whether
or
not
that
book's
worth
your
time
in
Reading,
okay
and
so
we're
going
to
add
a
feature
here
now
before
we
go
ahead
and
do
that
there's
a
few
things
that
we
ought
to
do
to
set
the
table.
B
I
think
the
first
thing
is:
let's
make
sure
that
this
repo
has
the
workflow
that
we
just
ran
in
the
other
project
inside
the
documentive
directory.
I.
Actually,
don't
think
it
does,
but
we'll
check
it
out
together
and
maybe
use
a
nice
Easter
egg
that
folks
are
less
familiar
with
in
order
to
introduce
it
before
we
go
ahead
and
Implement.
B
That
rating
feature
so
I'm
going
to
go
ahead
into
my
documentive
directory,
which
is
where
my
actions
workflows
will
live
and
see
if
my
AI
workflow
already
lives
here
and
if
it
doesn't
I
think
that's
probably
okay
but
yeah.
Let's
go
back
and
copy
this
workflow,
so
I'm
going
to
go
ahead
and
look
at
this
raw
workflow
I'm
going
to
copy
that
and
then
I'm
just
going
to
launch
this
lightweight
vs
code
editor
directly
from
my
repo
by
pressing
dot.
B
So
again,
what
should
happen
here
is
when
I
open
a
pull
request
against
my
main
branch.
A
Linux
machine
should
should
kick
off
a
you
know
the
following
steps,
which
is
to
check
out
the
latest
version
of
this
repo
and
uses
AR
AI
PR
integration
at
the
tag
version
at
B1.
Okay.
So
that's
the
task
at
hands:
let's
go
ahead
and
Commit
This
back
to
our
main
branch.
B
Something
awesome
is
a
foot,
so
let's
go
ahead
and
commit
and
push
that
and
I
think
we
could
actually
just
click
to
go
back
to
our
repo
I'll.
Go
ahead
and
close
out
these
other
repos
to
keep
things
nice
and
simple,
okay,
great!
So
now
we
should
have
that
workflow
if
I
go
back
to
dot,
GitHub
I
go
to
my
workflows,
I
see
AI
Galaxy,
which
is
the
AI
workflow
awesome.
Okay.
B
That's
been
deployed
on
your
behalf
that
has
Visual
Studio
code
installed
on
it
and
any
of
the
dependencies
of
your
application
that
you've
already
defined
in
order
to
run
that
application.
The
first
time
so
I
think
those
are
the
two
things
that
I
thought.
Oh
yeah,
of
course
also
your
source
code
of
your
repository.
So
those
are
really
the
three
things
that
your
code
space
gets,
gets.
B
You
know
a
higher
power,
laptop
I'm,
just
kind
of
utilizing
the
power
of
the
cloud
here
to
have
access
to
these
resources
and
what
you'll
notice
is
that
we've
actually
defined
so
we've
codified,
what's
required
to
run
this
application,
and
so
we
don't
allow
two
core
machines
for
this
specific
repo,
and
so
somebody
was
calling
out
like
Hey.
How
do
you
realize
that
level
of
maybe
velocity
the
speed
to
productivity?
B
B
So
now
we
have
this
vs
code
and
a
VM
somewhere,
and
maybe
the
first
thing
I
want
to
do
is
like
run
my
application
and
I
can
do
that
by
hitting
function,
F5
and
I'm
going
to
forward
the
application
to
a
port.
That
I
could
then
open
on
my
browser
here,
and
so,
let's
we're
running
this
application,
that's
being
forwarded
to
port
8080.
So
if
I
go
to
the
URL,
that's
getting
forwarded
here,
it's
going
to
say
connecting
the
port
beautiful,
okay,
so
here's
the
bookstore
deployed
or
not
necessarily.
B
Actually,
it's
not
deployed
it's
being
run
from
my
local
Dev
environment,
so
where
you
saw
it
before
was
deployed
to
Azure.
This
is
my
local
Dev
environment
right,
and
so
what
I
actually
want
to
do
is
pretty
simple
here.
What
I
want
to
do
is
run
a
piece
of
automation
prepared
for
this
demo.
That
allows
me
to
introduce
that
feature.
I
was
talking
about,
which
is
the
star
rating.
It's
not
going
to
spend
your
time
actually
implementing
that
with
you.
We
are
going
to
write
some
code
later
with
copilot,
so
fear,
not
okay.
B
B
B
B
The
whole
purpose
here
was
to
kind
of
show
you
the
developer
experience
in
the
workflow,
but
also
that
we
get
back
to
this
point
now
that
we're
introducing
a
feature
in
a
more
production-like
environment
to
see
whether
or
not
our
AI
integration
is
listening
right
that
we
introduced
just
a
moment
ago,
directly
back
to
the
main
branch.
So
we
might
say
something
like
I,
don't
know,
provide
a
rating
feature
in
the
bookstore
application.
B
So
we
know
whether
the
book
is
loved
by
its
readers
go
ahead
and
and
open
a
pull
request
there,
and
so
we
should
start
to
see
a
series
of
checks
get
kicked
off
beautiful,
so
it's
pretty
immediate,
so
there's
a
series
of
checks
getting
kicked
off
automatically
here
because
you
have
get
of
actions
in
the
background.
Listening
for
events
on
the
platform.
In
this
case,
it's
pull
request.
B
Events
and
the
good
news
is
I,
did
see
AI
test
here,
getting
kicked
off,
and
so
hopefully
that
resolves
in
a
moment-
and
we
should
see
the
result
of
that
back
in
rpr.
But
in
the
meantime
you
might
notice
like
this
pending
status
across
code
scanning
and
so
we're
also
scanning
spoiler
alert.
So
we're
also
scanning
the
auto
scroll,
spoiler
alert,
we're
scanning,
for
you
know,
appsec
type
vulnerabilities
with
code
ql
and
again
you
can
always
click
into
these
things
and
check
out
the
live
logs
as
they're
actually
running.
B
So
in
this
case,
codeql
is
getting
initialized
you'll
see
the
analysis
step
for
me
is
the
most
interesting
because
you
actually
see
the
queries
that
you're
running
against
your
application
in
real
time.
Whether
or
not
your
application
has
vulnerabilities.
So
that's
really
cool
okay,
so
I'm
going
to
scroll
up
here
and
just
show
you
what
you've
already
glanced
at,
which
is
this
pull
request,
is
about
introducing
a
bookstore
reading
feature
the
bookstore
application.
It
will
allow
customers
to
rate
books,
they've
read
which
could
help
other
customers
decide
what
book
to
purchase.
B
You
know
this
feature
could
also
help
the
book
store,
track
customer
preferences
and
make
recommendations
about
new
titles.
So
I
only
read
that
to
you,
because
it
was
the
first
time
I
was
reading
it
myself,
but
it's
it's
pretty
interesting
that
you
can.
You
know,
provide
such
limited
context
to
the
large
language
model.
B
It
will
synthesize
that
and
publish
it
back
for
you
now
one
thing
that
I'll
say
since
we're
here
now
and
since
we
kind
of
finished
the
demonstration
in
this
repository
is
this:
somebody
was
calling
out
copilot
x,
one
of
the
capabilities
under
the
copilot
X
hood
is
or
one
of
the
planned
capabilities.
Rather
that's
in
technical
preview
is
an
experience
that,
as
I'm
writing
a
pull
request.
Comment
like
that
opening
comment
you
saw
before
we
do
use
gpt4
and
more
advanced.
B
B
Okay,
I'm,
going
to
just
take
a
10
seconds
here
to
just
poke
back
in
the
chat
and
see
what
folks
are
we're
talking
about.
All
Tech
is
Magic.
Dan
I
couldn't
agree
more
seeing
some
folks
awesome,
Christopher
you're
the
best.
Thank
you
sharing
the
context
there.
No
love
for
lean
startups
automatically
is
not
in
correct
language
right
on
so
you're
all
having
a
little
bit
of
fun
talking
about
Automation
and
the
impact
of
some
of
these
tools
in
your
own
workflows,
so
I'm
grateful
for
that.
Thank
you
for
for
just
continuing
to
participate.
C
B
Yeah,
so
let's
talk
about
the
developer
experiences
for
the
developer,
who
actually
built
this.
This
Automation
in
the
first
place
right,
and
so
we
did
three
things
right.
We
started
out
just
testing
some
Automation
in
a
repo,
and
then
we
tested
that
into
more
production-like,
environment
and
I.
Think
the
third
thing
that
we're
going
to
do
together
is
actually
build.
This
integration
with
one
another
okay
and
check
out
the
developer,
experience
there
and
there's
a
lot.
That's
happening
implicitly
right.
B
B
B
Ahead
and
open
it
and
see,
what's
already
been
done
there
together,
so
it's
already
on
warm
standby,
so
you
saw
how
much
faster
that
was
and
I
do
see
some
of
the
files
open
here,
and
so
that's
okay.
As
long
as
there's
no
code
changes
because
I
want
to
do
those
code,
changes
live
with
you
all,
so
you
can
see
the
experience
watch
me
sweat
a
bit
and
you
know
and
again
it's
the
kind
of
the
magic
the
power
of
co-pilots
live,
so
everybody's
dropping
numbers
in
there.
That's
hilarious
and
fun
so
appreciate
you.
C
B
Know
actively
participating
there,
cool
okay,
so
Christopher
is
like
Kevin
you're,
destroying
the
chat,
everybody's,
just
spamming
numbers
now
and
I
see
Jesse
smiling
as
well.
Okay,
so
let's
go
ahead
and
do
this
together.
Let's
take
a
moment
to
maybe
just
acknowledge
how
this
action
manifested
as
that,
and
so
this
action
yaml
file
is
a
definition
of
an
action.
So
it's
actually
an
animal
file
is
a
definition
of
an
action.
So
pretty
simply
you
can
name
your
action.
You
can
give
it
a
description
and
you
can
provide
inputs,
which
is
obviously
incredibly
important.
B
We
use
the
open,
AI
key
import
input
for
something
obvious,
which
is
to
make
a
request
out
to
open
AI,
and
then
you
use
the
repo
token
to
take
the
synthesis.
That's
output
from
an
open,
Ai
and
plug
that
back
into
the
pr,
and
then
you
get
a
little
bit
of
context
about.
You
know
how
that
actions
actually
run
itself.
B
Okay,
so
let
me
go
ahead
and
do
this.
Let
me
just
make
sure
I
have
a
terminal
available
to
me.
Should
I
need
that
and
make
sure
there's
no
changes
here:
okay,
wonderful,
okay!
So
for
me
personally
for
me
personally,
one
of
the
things
I
see
this
is
Ezekiel
logo
down
here.
Okay,
so
now
it's
or
not
so
for
me
personally,
one
of
the
first
things
that
I
always
do,
especially
lately
when
I'm
doing
Dev
is
I'll,
always
make
sure
that
I
have
copilot
installed
in
this
case.
B
I've
already
installed
it
manually,
but
I
definitely
encourage
you
go
to
the
marketplace:
install
copilot.
If
you
have
access
to
the
service
before
excuse
me
before
you
begin
your
development,
so
we
can
go
ahead
and
go
back
to
the
files,
and
maybe
one
of
the
first
things
that
we
want
to
do
together.
B
I
forgot
to
show
it
to
you
in
the
other
repo,
but
we'll
talk
about
it
now,
one
of
the
first
things
that
you
may
want
to
do
when
you're,
maybe
adopting
code
spaces
for
the
first
time
or
creating
a
project,
is
again
codifying
that
project's
dependencies
such
that
you
can
run
the
application
really
easily.
You
know
the
first
time
you
load
it
so
I
don't
have
to
come
over
to
your
desk
and
be
like
hey
man.
What
are
all
the?
B
What
are
what
is
this
crazy
error
message
that
I'm
getting
and
you
respond,
and
you
say
something
like
oh
well,
you
have
to
install
this
thing
that
I
remember
because
I
joined
three
years
ago-
and
somebody
told
me
when
I
joined,
then
you
know
like
that
problem
kind
of
goes
away
when
you
codify
the
dependencies.
Which
is
less
job
security,
but
obviously
a
better
experience
for
Developers
cool.
B
So
let's
go
ahead
and
add
a
customization
I.
Think
in
this
case,
if
I'm
not
mistaken,
I
want
to
do
something
like
add
vs
code
and
I.
Think
I
want
to
just
make
sure
that
I
have
okay.
So
this
is
interesting.
So
copilot's
already
looking
over
my
shoulder
and
providing
suggestions
for
what
I
might
like
to
see
in
this
vs
code.
B
Yeah
object
here
in
this
case
it's
extensions
and
the
docker
file
definition
for
me,
I'm,
going
to
keep
it
pretty
light
today,
so
I'm
just
going
to
have
extensions
there,
I,
don't
know
how
many
times
I've
done
that
extensions
and
I'm
going
to
add
a
column
there
and
then
pass
this
array.
In
this
case,
it's
just
going
to
be
GitHub
I.
B
B
Similarly,
I
might
have
something
that
you
know
reads
my
package
management
system
and
allows
me
to
to
actually
execute
that
when
I
open
the
code
space
so
fun
times,
definitely
encourage
you
to
leverage
the
dev
container
capability
and
code
spaces
is
a
really
nice
kind
of
accelerator
for
getting
productive
or
being
productive
or
just
making
changes
on
the
Fly
to
Your
app.
So
that's
really
useful.
B
Okay,
now
I
realize
it.
We
have
about
five
to
seven
minutes
left
together
in
the
demo
portion,
then
we're
going
to
open
up
for
I,
think
q,
a
and
open
discussion,
so
I'm,
just
kind
of
checking
in
with
you
all
on
where
we
are.
We
have
approximately
five
to
seven
minutes
to
implement
some
features
together
and
then
we
will
open
up
for
Q
a
so.
Let's
see
some
folks
in
the
chat
here,
where's
my
hoodie
Sonora,
Lee
and
I
see
some
folks
personally.
Keep
using
copilot
is
copilot
available
and
other
SCM
tools.
B
Ai
upscaling
AI
remembers
our
comments.
Wonderful,
okay,
so
great
conversation
happening
in
the
background,
all
right
friends.
So
let's
do
this,
don't
get
I
know.
Y'all
are
developers
so
you're
not
intimidated
by
code
generally,
but
just
don't
get
intimidated
about
the
idea
that,
like
you,
have
to
read
the
code
to
understand
what
the
heck's
happening
here
all
I
wanted
to
to
show.
You
was
what's
happy
on
line
six.
Is
you
know
because
it's
an
action?
You
basically
have
this,
try
catch
block
so
there
it
is.
B
C
B
Okay,
I'm
in
the
wrong
file,
I
was
wondering
what
is
going
on
there.
You
have
the
Stray
catch
block.
It
defines
the
logic
of
what
happens
in
your
action
to
keep
it
super
simple.
We
get
the
action
event,
and
here
it
is
we're
just
kind
of
like
capturing
that
stuff
assigning
at
the
variables
that'll
be
available
to
me
and
now
what
I
want
to
do
is
when
I
have
the
output
of
my
PR.
B
B
So
what
you
might
have
noticed
was
that
my
inline
co-pilot
was
already
beginning
to
suggest
implementations
here,
and
so
what
I
did
was
I
hit.
Ctrl
enter
that
opened
up
this
side,
tab
that
describes
to
me
the
different
available
options,
and
so,
let's
check
out
the
first
one
here,
it
says:
const
response
equals
opening
eye
create
completion.
It
calls
it
has.
You
know
all
the
attributes
that
you
need
for
a
successful
request
to
open
AI,
okay
using
DaVinci.
It
says
so.
B
This
is
the
magic
under
the
hood
by
the
way
like
if
you
want
to
see
a
way
that
just
I
don't
know
if
it's
clever
or
not-
probably
not
very
clever,
but
in
my
dull
mind
it
was
what's
this
pull
request
about
and
how
can
impact
my
application?
Here's
the
pr
title
with
this
PR
body.
So
that's
as
simple
as
it
is
I,
just
kind
of
convert
the
out
the
input
of
a
PR
into
plain
text
and
then
send
it
off
to
the
large
language
model
for
synthesis
and.
B
B
It
would
also
ask
me
to
First
Define
that
as
a
dependency
in
my
app
and
then
it
would
allow
me
to
call
it
so
I'm
going
to
go
ahead
and
accept
this
solution
right
on
so
I
see
that
and
now
the
second
thing
that
I'm
going
to
do.
The
second
and
last
thing
actually
that
I
want
to
do
is
I
want
to
call
the
Octo
kit
SDK,
so
the
GitHub
SDK,
to
take
the
output
of
from
openai
and
post
that
back
in
the
pr
and
so
I'm
just
going
to
do
my
inline
tabbing
here.
B
So
this
is
typically
my
experience
as
I'm
doing
my
development
is
like
I
just
hit
tab
tab
enter
as
I'm
expressing
what
I'd
like
to
see
implemented
and
I
obviously
have
familiarity
with
what
a
a
reasonable
implementation
looks
like,
and
so
this
looks
reasonable
to
me.
We're
gonna
it's
gonna.
This
function
is
asynchronous
and
it's
going
to
comment
back
in
my
PR
as
the
really
it's
the
as
get
a
git
of
action,
and
it's
just
going
to
post
back
the
comments
which
comes
earlier
from
from
this
create
completion.
B
B
B
Oh,
did
anyone
guess
one?
Were
there
any
ones
I
doubt
it
I
think
we
I
saw
a
lot
of
42s,
so
we're
gonna
go
back
in
the
chat
there
and
see
if
there's
any
ones
Carlos
I
saw.
You
were
first
but
yeah
a
little
too
late
there,
my
friend,
okay,
so
let's
go
ahead.
I
kept
true
to
my
word,
I'm
glad
I.
Remember
that
and
let's
check
the
good
status.
Okay
at
all
gets
commits
complete
integration.
B
Does
anyone
remember
look
for
the
chat
here,
Sam
giving
me
some
tips?
Thank
you,
sir.
Does
anyone
remember
what
the
original
Branch
was
called
live?
Okay,
great,
so
I'm
gonna
go
ahead
and
enroll
with
that,
then
push
back
to
the
main
branch
go
ahead
and
commit
okay,
and
now
we're
going
to
open
a
pull
request
again
so
I'm
going
to
go
back
and
just
show
you
what
that
experience
is
I'm
going
to
scaffold
out.
B
B
And,
let's
see
if
we're
able
to
make
that
change
successfully,
so
the
pr
should
trigger
the
workflow
and
then
let's
track
this
workflow
executing
our
live
logs
here.
What's
interesting,
you
also
see
the
Prisma
Cloud
scan.
So
it's
looking
for
infrastructures
code
I,
don't
have
any
infrastructure
as
code
in
this
repository.
So
here
are
the
live
logs.
What's
really
nice
is
you
saw
it
like
because
I'm
using
the
native
secret
Square,
we
automatically
redact
the
secrets
out
of
the
out
of
the
workflow
logs.
So
never
like
find
my
secret
output
there.
B
Let
me
go
back
to
the
pr
see
if
something
failing,
but
our
integration,
Works
friends,
so
congratulations.
This
pull
request
introduces
AI
integration
in
your
app
and
it
helps
detect
security,
risks,
etc,
etc.
So
that's
it
friends.
That's
all
I
had
to
share
with
you.
Thank
you
so
much
for
your
time
and
attention
I
had
a
ton
of
fun
doing
it,
and
thank
you,
of
course,
to
Jesse
and
Christopher
for
moderating
the
chat
and.
D
Thank
you
both
for
for
everything
that
was,
that
was
fantastic,
so
there's
been
a
ton
of
chat
and
a
ton
of
q,
a
as
as
the
both
of
you
have
have
seen.
I
want
to
start
with
by
far
I
think
the
most
important
question,
especially
since
we
had
that
light
mode.
Dark
mode
debate
is
GIF.
How
do
we
pronounce
that.
B
D
C
D
B
Right
on
so
one
of
the
things
that
I
did
to
just
illustrate
the
answer,
because
I
tend
to
be
a
little
long-winded,
I
apologize
in
advance,
one
of
the
things
that
I
was
doing
in
order
to
ensure
that
co-pilot
had
a
very
high
probability
of
success
was
I
opened
up.
Another
file
that
actually
had
a
definition
of
the
functions.
I
was
writing
in
the
other
file,
and
so
that's
the
answer.
B
So
it
looks
forward
for
context
into
your
applicant
into
your
source
code
file
and
it
looks
backwards
into
what
you've
already
written
in
order
to
increase
the
probability
that
its
suggestion
is
going
to
be
accepted,
which
is
kind
of
the
key
metric
that
we
use
to
track,
whether
or
not
there's
a
quality
output
which
is
acceptance,
the
acceptance
rate,
okay
and
it's
not
your
source
code,
so
we're
not
training
the
model
on
your
source
code.
Yes,
that
I
think
that's
like
a
key
Point
make
sure.
Folks
know,
thank
you,
yeah.
D
That's
always
one
of
the
biggest
questions
that
I
get
asked
so
Jeremy.
Thank
you.
Thank
you
for
that.
I
also
want
to,
and
I'll
probably
I'll
find
the
name
here
in
a
minute,
but
pivoting
real
quickly
to
code
spaces.
So
one
of
the
things
that
you
demoed
is
that
you
that
you
made
a
change
to
the
service,
and
then
you
were
able
to
like
open
that
up
in
a
browser
to
be
able
to
like
actually
see
it
and
test
it
and
so
forth.
D
A
Yeah,
so
you
have
controls
over
that
for
for
the
Enterprise
right.
So
by
default
it's
going
to
be
public
or
private,
and
you
do
have
the
ability
to
set
up
some
forwarding
so
that,
if
you
would
like
to
have
it
able
to
share
just
to
people
who
have
access
to
that
repository,
you're
also
able
to
do
that
Kevin
anything
else.
You
would
want
to
add
to
that
I.
D
Cool-
and
that
was
a
breath
that
I
asked
asked
that
question
I
now
want
to
go
back
to
copilot
here,
real,
quick
and
I
apologize
for
kind
of
bouncing
back
and
forth.
There's
so
many
questions
that
have
been
asked
been
fantastic,
but
going
back
to
co-pilot.
One
of
the
other
questions
that
had
come
up
is:
is
it
possible
to
train
co-pilot
on
our
code
and
be
able
to
like
show
it
our
our
code
base?
So
we
can
get
a
better
sense
of
what
it
is
that
we're
doing
and
how
we
do
it.
B
So
yeah
so
I
think
what
yeah
so
I
think.
The
answer
here
is
and
probably
what
everybody's
like
anxious
for
me
to
get
to
if
I
may
is
so
we
we
are
working
on
the
the
ability
to
increase
so
to
train
the
model
to
allow
you
to
bring
your
own
source
code
to
influence
the
model's
output.
It's
not
something
that's
available
today
and
it
would
be
obviously
explicitly
opt-in
by
organizations
whether
or
not
they
want
their
own.
B
Now,
I
guess,
and
if
you
wanted
something
like
a
junior
developer,
to
have
access
to
like
gold,
standard
or
Boiler
boilerplate
code,
and
you
don't
have
the
ability
to
train
a
custom
model
yet
there's
an
idea
that
you
might
be
able
to
like
open
up
a
separate
file
that
has
all
those
gold
standards
defined
and
allow
that
to
influence
the
context
of
your
suggestions.
I'd
love
to
hear
what
this
group
like
thinks
about
that
idea.
D
Cool
I
also
do
want
to
add
in
a
real
quick
plug
for
GitHub
next.
So,
if
you
go
there,
you
can
actually
see
some
of
the
different
things
that
are
being
planned
as
part
of
copilot
X
and
in
the
update
upcoming
features
there
as
well.
So
definitely
go
check
out.
Github
next,
just
go
to
your
favorite
search
engine.
D
Get
up
next
and
you'll
be
able
to
find
that
find
that
there
going
back
to
projects,
because
I
want
to
make
sure
that
we
also
pay
a
little
bit
of
attention
to
projects
as
well.
Vincent
asked
a
question:
I'm
gonna
kind
of
slightly
put
a
spin
on.
It
is
what
do
you
suggest
for
templating?
So
let's
say
that
you
know:
we've
got
five
or
six
projects
that
we're
all
going
to
be
spinning
up
they're
going
to
be
relatively
similar
like.
How
do
we
then
template
that?
D
Let's
see,
I'm
kind
of
flipping
back
gersha
may
ask
a
great
question,
which
is
one
that
I'm
seeing
popping
up
quite
a
bit
and
I'm
again
going
back
to
copilot,
everybody
loves
co-pilot
but
I'm
again
going
back
to
co-pilot,
which
is
what
are
some
tips
for
getting
the
most
out
of
it.
That
obviously
this
is.
This
is
a
new
tool.
D
D
The
next
is,
if
co-pilot
isn't
doing,
what
you
would
expect
it
to
do
is
maybe
go
back,
especially
if
you're
doing
it
like
with
a
comment.
So,
let's
say
I'm
I'm,
saying:
hey,
create
a
class
that
does
X
Y
and
Z,
and
it
generates
some
some
code
for
me
and
I.
Look
at
and
go.
That's
not
quite
right.
Go
back
to
your
comment
and
just
redefine
or
like
add
on
a
little
bit
of
extra
context,
to
kind
of
help.
D
Point
co-pilot
in
the
right
direction
that
if
co-pilot,
has
way
too
much
room,
it
doesn't
white
know
what
to
do,
and
sometimes
the
suggestions
are
going
to
be
relatively
generic.
If
you
trim
it
down
a
little
bit
pare
it
down
a
little
bit,
then
it's
going
to
be
able
to
to
follow
along
and
then
the
last
big
thing
that
I
would
say-
and
this
kind
of
goes
to
to
Kevin's
little
like
behind
the
scenes
trick
is
copilot
will
watch
what
you're
doing,
and
it
will
make
suggestions
based
on
that
now.
D
D
No
so
I
wear
this.
Just
for
this
is
my
Apple
watch
and
I
wear
this,
for
you
know
just
like
normal
Fitness,
stuff
and
I
wear
this,
because
I
need
some
level
of
style.
This
is
a
a
Swatch
back
when
I
was
a
kid.
These
were
really
really
popular
and
I
couldn't
afford
them.
When
I
was
a
kid
and
now
that
I'm,
an
adult
I
can
afford
them
so
yeah.
So
this
is
actually
the
Berlin
specific.
A
We
got
we
got
a
question
about
you,
had
your
your
bookmarks
bar,
showing
when
you
were
demoing,
and
you
had
that
mono
repo
bookmark
up
there,
so
somebody
wanted
to
talk
about
monorepas
I
thought
that
was
a
decent
thing
to
maybe
bring
up
in
the
context
of
what
you
were
working
on.
If
you
had
like
you
know,
one
thing
to
say
about
Mono,
repos
and
developer
velocity.
B
B
Yeah,
well,
you
can
also
yeah
so
that
maybe
that's
true
but
I
think
for
our
new
Services.
We
have
adopted
more
of
a
micro
service
architecture
and
I've
kind
of
broken
some
of
that
stuff
out,
because
it's
created
a
lot
of
overhead
like
when
we
started
talking
about
the
old
experience
developers
had-
and
this
is
not
necessarily
to
your
that
point
specifically
but
like
it-
used
to
take
10
hours
to
get
productive
with
GitHub
and
now
you
know
with
so
many
levels
of
refinement,
including
I,
would
imagine
re-architecting
the
application
itself.
A
Three
builds
and
that's
kind
of
what
you
were
hinting
out
there
right
pre-builds
are
godsend.
The
larger
your
code
code
base
gets
the
more
complex
your
code
base
gets
being
able
to
have
that
that
builds
it
and
waiting
in
the
wings
so
that
you're,
not
spending
so
much
time.
Bootstrapping
really
can
be
helpful.
That's
something
we
rely
on
heavily
here.
A
D
There's
another
question
that
just
came
in
here
and
this
isn't
necessarily
like
an
overly
technical
question,
but
it's
something
I'm
willing
to
bet
that
there's
a
fair
amount
of
the
audience
that's
interested
in
and
I'm
actually
going
to
expand
upon
it.
So
Brad
didn't
ask
the
question:
is
it
possible
for
coding
to
become
like
an
entirely
like
automated
profession
and
then
at
a
higher
level
like
what
what
advice
for
like
somebody
wanting
to
become
a
developer,
or
would
you
give
them
in?
Would
you
give
them
today.
A
No
I
I
don't
think
we
can
ever
automate
away
the
the
Judgment
capacity
of
humans
right
you,
you
still
are
an
expert
on
something
or
you
are
becoming
an
expert
on
something,
like
you
know,
prove
me
wrong.
That's
great,
but
I
think
there's
still
like
I
kind
of
mentioned,
the
idea
of
getting
to
focus
on
the
business
Logic
on
what's
really
important
for
customers.
A
Instead
of
just
doing
the
boilerplate
things,
I
think
that's
really
what
the
focus
here
is
not
like
making
development
a
thing
of
the
past
and-
and
we
all
could
do
different
jobs
right
and
I-
think
that
ties
in
really
nicely
to
the
idea
of
how
do
I
learn
and
grow
as
a
developer,
using
something
like
copilot,
because
it
can
really
help
you
explore
things
in
an
accelerated
way
that
you
otherwise
before
would
have
had
to
be
stuck
in
the
docks.
You
know
doing
a
lot
of
training
right.
A
You
can
just
kind
of
jump
in
and
start
playing
around
and
yeah.
Maybe
you're
still
going
to
make
mistakes,
but
you
get
a
little
bit
of
that
like
fail
fast
right,
where
you
can
get
suggestions
play
around
with
it
and
start
trying
things
out
without
having
to
be
like
I
said
so
stuck
in
the
docks
and
the
the
really
learning
piece
of
it
and
Kevin
I,
don't
know
if
you
have
anything
else,
you
would
want
to
add
to
that.
A
D
Can
learn
a
lot
perfect,
I
I
think
that's
a
perfect
note
to
to
to
end
on
so
Kevin
Jesse.
Thank
you
so
very
much
for
everything
that
that
you
shared
today,
just
a
real,
quick
glanton
in
the
chat
shows
that
everybody
I
love
the
the
session.
So
take
a
minute
and
kind
of
go.
Go
see
what's
been
going
on
in
there
to
everybody
that
tuned
in.
Thank
you.