►
Description
Slide deck: https://docs.google.com/presentation/d/1uU7wEd0E0hOxdK-TfR1a3OmzNOHHkzRnQ84hiTcIejQ/edit?usp=sharing
00:09 Why Automation
01:09 Why automation is hard?
01:39 What is No-code Automation?
02:45 Our Vision
03:45 UX Preview
04:50 Architecture
05:50 Execution Strategy
07:02 Demo
09:04 Progress & ETA
09:29 Appendix
A
Hi
folks,
I'm
tell
from
gear
Labs
no
Code
and
low
code
single
engineer
group
today:
I'll
showcase
the
local
automation,
new
product
direction
that
I've
recently
chosen
to
productionize.
So
why
automation,
regardless
of
your
role
as
a
professional,
we'll
log
on
to
a
computer
every
day
in
the
morning,
you're
most
likely
working
on
for
cure
task,
whether
it's
from
your
to-do
list
emails
like
messages,
tickets,
merge
or
pull
requests?
Of
course,
many
of
the
tasks
require
creative
thinking,
but
there
are
also
tons
of
them
that
are
Fields,
be
mechanical
and
repetitive.
A
Just
imagine
the
time
you
can
say
by
automating
some
of
those
boring
tasks
so
coming
out
of
the
pandemic.
The
increasingly
distributed
and
asynchronous
Workforce
also
means
that
less
tolerance
of
the
time
wasted,
while
waiting
for
simple
actions
to
take
place,
such
as
assigning
a
user
when
a
label
is
added
to
an
issue
if
you're
already
familiar
with
gitlab,
you
understand
the
continuous
software
delivery
or
CI
CD
is
essentially
a
formal
automation,
accept
their
design
for
developers
to
manage
their
code.
A
Business
user
would
love
to
automate
their
tasks
too,
but
automation
is
not
easy,
they're
unique,
which
means
and
there's
no
one-size-fit-out
solution
from
Team
to
team
business
to
business.
They
all
work
differently
and
it
requires
uniquely
codified
rules
to
automate
and
they're.
Also
complex
many
workflows
involve
multiple
business
functions
across
multiple
systems,
they're
costly
too.
They
require
not
only
the
coding
skills
to
build
and
maintain
and
operate,
but
also
process
tend
to
change
from
time
to
time.
A
So
there's
no
one-off
investment
there,
and
this
is
where
the
local
automation
comes
to
rescue
local
automation
tools,
allow
users
to
define
a
manual
rule-based
programs
that
automate
tasks
without
writing
a
single
line
of
code.
So
what's
significant
here
is
a
self-service
model.
The
no
Cloud
2
helps
business
users
to
become
programmers
themselves,
and
then
they
can
teach
computers
to
do
their
job
on
behalf
of
them.
It
achieves
that
by
replacing
the
programming
language
that
is,
take
ages
to
learn
and
Perfection,
with
a
user-friendly
and
intuitive
graphical
interface,
with
very
little
learning
curve.
A
Jira
automation,
for
example,
that
launched
three
years
ago,
has
reached
10
billion
executions,
so
our
vision
is
a
gitlab
native,
no
code
2
for
business
users
to
automate
and
orchestrate
Parts
across
the
entire
devsec
Ops
lifecycle,
and
the
keywords
here
are
the
automated
orchestrator.
We
want
to
automate
simple
tasks
that
the
UI
base
2
is
able
to
handle,
while
also
the
more
complex
hard
to
the
more
capable
runtimes
and
professional
programmers
such
as
site
pipelines
where
the
code
shines.
A
If
you
look
at
the
competitive
landscape,
zero
automation
is
very
UI
Centric
by
the
nature
of
our
license
platform.
Of
course,
GitHub
actions,
on
the
other
hand,
are
popular
amount
developers.
Given
its
code
driven
approach
and
the
associate
flexibility,
we
want
to
bring
the
best
of
the
both
the
Both
Worlds
together.
We
invested
a
hybrid
model
where
the
local
automation
is
capable
of
completing
simple
tasks
without
writing
a
single
line
of
code,
at
the
same
time
able
to
orchestrate
code
executed
from
other
gitlab
capabilities.
A
I
I
also
want
to
give
you
a
sneak
peek
into
the
ux
design
that
we
have
so
far
on
this
screen.
You
can
see
we
have
the
trigger
condition
and
action
constructs.
You
don't
have
to
learn
how
this
ux
works,
but
we
want
to
make
it
kind
of
intuitive.
Just
by
reading.
Through
those
descriptions
you
can
see,
this
automation
rule
gets
triggered
when
an
issue
is
created
and
the
conditions
are.
We
have
three.
A
A
You
can
also
see
the
the
page
that
managed
the
automation
rules
and
their
status
and
execution,
and
the
logs
and
and
everything
so
I
want
to
give
a
special
shout
out
to
the
the
designer
and
the
vote
for
helping
out
refining
the
ux.
So
we
also
have
architectural
design
I
put
into
place.
It's
quite
conventional
and
you
can
also
say
it's
a
bit
boring.
We
have
the
standard
front
and
the
back
end
architecture.
In
the
front
end,
we
have
two
apps
one
focus
on
the
building
experience.
A
A
We
also
have
the
cache
layer
in
postgres
or
radius
so
and
those
automation
get
triggered
by
the
the
hooks
when
starting
bands
occurs,
and
then
we
put
them
into
a
queue
and
for
the
worker
to
pick
up
and
then
and
run
those
rules,
they're
able
to
interpret
the
dsls
and
and
just
take
action
from
there.
So
with
all
this
design
in
place,
let's
talk
about
the
execution
strategy.
We
want
to
start
with
project
management
such
as
automating
the
tasks
around
the
issues
and
epics.
A
This
is
a
business
domain,
mostly
owned
by
the
non-developer
users,
which
means
that
they're
going
to
benefit
the
modes
from
the
no
code
tools.
We
also
want
to
focus
on
the
ux
for
both
user
friendliness
and
extensibility.
This
is
important
because
our
goal
is
to
give
everyone
the
opportunity
to
automate
their
repetitive
tasks,
regardless
of
their
technical
background,
the
extensibility
part
of
the
framework,
and
what
is
that
right?
It
can
make
the
expansion
into
other
areas
that
that
SEC
Ops
lifecycle
a
lot
easier
down
the
track.
A
A
So
let
me
quickly
replay
the
demo
that
I
recorded
in
the
previous
update
for
those
who
hasn't
seen
the
the
demo.
Yet
imagine
we
want
to
build
an
automation
that
notifies
the
product
manager
with
a
nuclear
issue
has
received
more
than
two
columns,
so
you
can
see
from
the
screen.
The
automation
is
broken
down
into
three
sections,
with
a
trigger
we're
going
to
select
issue
and
the
condition
with
date.
We
want
to
make
sure
it's
a
new
issue
one
day
and
also
the
issue
has
no
labels.
Yet.
A
A
A
A
You
look
at
the
label
that
here
is
a
new
label
edit
and
now
we
have
a
notification
center,
better
automation,
you
forgot
to
mention.
If
you
have
some
idea,
you
want
to
share
with
me.
Please
leave
the
comments
either
here
or
find
the
link
in
the
description
below.
Thank
you
very
much
for
your
time
and
see
you
next
time.
A
So
in
terms
of
the
progress
and
ETA
I
would
say,
we've
already
made
a
series
of
hard
decisions
in
terms
of
design
and
architecture
and
right
now
we're
knee
deep
in
the
the
implementation
phase,
with
a
few
imrs
being
worked
time
parallel
and
a
couple
of
them
already
merged.
Just
look
at
the
back
clock.
It
looks
like
it's
going
to
take
perhaps
two
to
three
months
to
complete
the
first
card
of
the
MVC.