►
Description
Presented by Abi Noda, Founder & Developer (Pull Reminders).
Want to build a profitable side project? In this session I will share the story of how my GitHub App that began as a side project became a successful business used by over 500 companies like Pivotal, WeWork, and Trivago. You'll learn about how the GitHub Marketplace is empowering independent developers to become entrepreneurs and tips for how you can build your own GitHub Apps for fun or profit.
About GitHub Universe:
GitHub Universe is a two-day conference dedicated to the creativity and curiosity of the largest software community in the world. Sessions cover topics from team culture to open source software across industries and technologies.
For more information on GitHub Universe, check the website:
https://githubuniverse.com
A
A
A
A
At
the
beginning
of
January,
I
flew
to
Chicago
to
be
with
close
friends
and
family.
It's
freezing,
cold
and
I
was
in
a
dark
place.
I
was
completely
preoccupied
with
trying
to
figure
out
what
to
do
with
my
life.
I
thought
about
all
the
options
thought
about
just
getting
an
easy,
remote
job
and
traveling
the
world
thought
about
trying
to
cash
in
and
move
to,
San
Francisco
and
get
a
lucrative
job
I
even
considered
a
career
change
and
looked
in
the
data
science
courses.
A
A
My
answer
came
to
me
instantly.
If
I
didn't
need
to
worry
about
money,
I
would
go
build.
My
own
thing
build
my
own
product
and
work
for
myself
and
in
fact
this
had
always
been
my
dream
lurking
in
the
back
of
my
mind,
behind
fears
of
failure
and
not
having
money,
but
in
this
moment
I
thought
I
could
do
it,
but
if
I
got
certain
now
and
didn't
quit,
eventually
I
could
succeed,
and
so
I
did
in
the
next
four
months.
A
A
So
when
I
figured
out
what
I
was
gonna,
you
know
do
with
my
life.
The
first
thing
I
had
to
do
was
figure
out
how
I
was
gonna
pay.
My
bills
I
had
no
income
and
I
really
like
buying
the
organic
avocados
at
the
grocery
store,
and
you
know
I
needed
some
way
to
cover
everything
and
I
also
knew
that
I
wanted
to
give
myself
the
best
chance
to
succeed,
and
in
order
to
do
that,
I
had
to
get
myself
a
long
runway.
A
I
don't
want
to
be
stressed
about
money
or
worried
about
you
know,
paying
rent
and
so
I
actually
thought
about
getting
a
full-time
job
but
decided
against
it,
and
the
thing
was
I
wanted
to
be
able
to
cover
my
bills,
but
I
didn't
want
to
be
too
comfortable.
I
wanted
to
be
on
my
toes
and
pressure
to
actually
come
up
with
an
idea
that
generated
money,
so
instead
I
applied
to
some
consulting
agencies
and
was
able
to
find
some
contract
work
right
away.
A
A
Remember
that
side,
project
I
told
you
about
I
started
working
on
after
I
got
fired.
That
was
poor
reminders
and
I'll.
Tell
you
a
little
bit
about
the
back
story,
so
the
job
I
got
fired
from
I
was
in
charge
of
managing
a
distributed
team
and
we
had
a
code
review
process.
That's
probably
similar
to
yours.
We
would
open
up
pull
requests
on
github,
assign
reviewers
and
share
them
in
slack
to
notify
the
team.
Raise
your
hand
if
this
sounds
similar
to
your
code
review
process.
Okay,
so
you
guys
probably
know
what
I'm
talking
about.
A
But
you
know
we
would
have
this
problem
where
you
know
people
would
share
their
pull
requests
in
slack
and
then
we'll
just
be
crickets,
and
you
know
you
would
think
of
people
are
probably
out
for
lunch.
You
know
busy
in
a
meeting
and
so
you'd
go
out
for
lunch,
get
some
bibimbap
or
whatever,
and
and
when
you
came
back,
you
just
don't
have
a
code
review,
and
so
this
was
our
problem.
A
You
know
code
reviews
would
take
anywhere
from
days
to
even
weeks
to
complete
and
so
a
feature
that
took
someone
one
day
to
build
could
take
weeks
actually
release
the
customers
because
of
our
house.
Well,
the
code
review
process
was,
and
so
from
that
point
from
that
point
on
I
started
spending
a
lot
of
my
own
time.
You
know
checking
up
on
pull
requests
and
following
up
with
people
who
needed
to
take
action,
I
hated
spending
time
to
slay.
A
My
advice
is
to
pick
a
problem.
You
personally
have
I
think
that
the
best
ideas
come
from
scratching
around
edge,
because
when
you
do
that
you're
intimately
familiar
with
the
problem
you're
trying
to
solve
and
also
who
you're
trying
to
solve
it.
For
so
I
was
tink
tinkering
with
the
side
project
pull
reminders-
and
you
know
at
this
point,
I
was
doing
some
contract
work.
A
I
was
able
to
buy
those
organic
avocados
and
I
wasn't
sure
if
I
should
continue
with
full
reminders,
because
I
wasn't
sure
if
it
was
something
that
anyone
to
actually
pay
for
or
want
to
use,
and
the
thing
is
like.
Maybe
a
lot
of
you
I've
had
a
lot
of
side
projects
in
the
past
and
pretty
much
everything
has
turned
out
to
be
something.
No
one
wants
to
use,
and
you
know
no
one
cares
about,
and
so
I
didn't
want
pull
reminders
to
suffer
that
same
fate.
A
Yeah
I
wanted
to
build
something
people
wanted,
and
so
I
did
a
couple
of
things.
One
thing
I
did
was
go
on
Google
and
just
search
for
github
pull
requests
for
mine
there
and
a
bunch
of
open
source
projects
came
up.
I,
remember
having
mixed
feelings
about
this.
On
one
hand,
I
was
like
yeah
I'm,
not
the
only
crazy
person
with
this
idea,
but
I
was
also
worried.
Cuz.
They
were
already
these
free
options
and
I
wasn't
sure
if
anyone
would
pay
for
mine
or
even
sign
up
for
it.
A
A
Typically,
you
can
control
things
like
frequency
or
you
know,
do
filters,
and
the
other
thing
was
that,
because
these
were
open-source,
it
would
take
quite
a
bit
of
effort
to
go
through
the
documentation
figure
out
how
to
host
that
set
it
up
and
run
it.
So
I
thought
to
myself
that
if
I
could
come
up
with
something
that
was
easy
to
use
and
simple
to
set
up
that
I
might
have
a
chance.
A
Something
else
I
did
was
I
went
on
the
Chicago
I'm
in
the
Chicago,
CTO
slack
group,
and
so
I
asked
people
hey.
Does
anyone
else
have
this
problem
of
PRS
taking
forever
to
review
and
I
got
a
pretty
good
response
and
in
fact
there's
even
someone
in
that
group
who
had
built
his
last
company
built
their
own
Bob
to
solve
this
same
problem.
A
So
at
this
point,
I
was
still
pretty
skeptical.
That
poor
reminders
could
make
any
money
or
be
a
business,
but
I
was
pretty
confident
that
it
was
a
real
problem
and
you
know
someone
would
want
to
use
that
and
I
think
that's
the
thing
with
ideas.
You
know
if
you're
overly
optimistic
you'll
go
off
and
you
know
build
whatever
idea,
you
have
and
no
one
want
to
use
it.
A
So
it
took
me
three
weeks
to
build
the
first
version
of
pull
reminders
and
when
I
first
started
working
on
it,
I
wanted
to
build
it
with
a
lick.
Sir,
for
those
of
you
don't
elixirs,
this
sort
of
new
fun,
programming,
language
and
I
was
really
excited
to
learn
it,
but
instead
I
built
pull
reminders
with
rails,
which
you
know
many
people,
maybe
not
in
this
room-
consider
to
be
old
and
boring
and
I
did
that,
because
I
thought
that
if
poor
reminders
was
potentially
gonna,
be
a
business
I
didn't
want
to
be
bogged
down.
A
You
know
spending
my
time
figuring
out,
elixir
stuff,
because
I
would
have
so
many
other
things
to
worry
about.
So
you
know
if
you're
trying
to
build
a
product
that
you
think
might
be
a
business,
my
recommendation
would
be
to
choose
the
most
boring
tool
as
possible.
It's
it's
easy
to
get
tempted
to.
You
know,
kill
two
birds
with
one
stone
and
learn
something
new
while
you're
building
your
product.
But
if
it's
gonna
become
a
business,
you
don't
have
time
to
be
dealing
with
tech
problems.
Then
you
know.
A
A
A
One
interesting,
interesting
part
of
building
pull
reminders
has
been
figuring
out
how
to
support
the
different
workflows
on
github.
So
when
I
first
started,
I
focus
on
what
I
thought
was
the
most
common
workflow,
which
is
to
open
a
pull
request
and
assign
individual
users
to
review
your
code,
but
as
time
went
on,
I
learned
that
there
are
many
different
workflows
on
github.
So
you
know
some
people,
instead
of
assigning
users
to
review
their
code,
will
assign
github
teams
and
other
teams
don't
use
review
requests
at
all.
A
They'll
use
things
like
labels
or
github
x'
required
review
settings.
You
know,
make
sure
their
pull
requests
get
reviewed,
and
so
you
know
what
I
learned
is
that
github?
Is
this
huge
platform
with
many
different
use
cases,
and
so,
if
you're
trying
to
build
a
github
app,
you
know
for
the
masses,
try
to
focus
on
the
most
common
use
cases
and
in
general
today,
I
think
people
are
inundated
with
new
tools
and
products
and
have
really
short
attention
spans.
A
I
got
a
few
likes
and
a
couple
of
my
friends
reached
out,
say
you
know
cool
app,
but
no
one's
seem
that
excited
about
poor
reminders,
and
you
know
at
this
point
I
still,
then
I'd
sue
only
think
poor
reminders
could
be
a
business,
and
so
my
advice
to
you
is
you
know
when
you
launch
your
product,
don't
get
discouraged
if
you
don't
get
a
billion
retweets
or
you
get
featured
on
product
ton
to
get
like
five
votes.
I
think
that
success
doesn't
usually
come
from.
You
know
one
tweet
or
one
blog
post.
A
A
So,
as
I
just
mentioned,
when
I
launched
poor
reminders,
no
one
seemed
that
excited
about
it
and
I
still
didn't
think
they
could
make
any
money
or
be
a
be
a
real
business
that
changed
when
I
got
an
email
from
a
guy
named
David.
So
David
is
an
engineering
manager
at
a
company
called
Monday
comm
and
when
I
got
an
email
from
him,
I
was
super
excited.
This
is
the
first
time
that
a
real
person
from
a
real
company
had
expressed
interest
in
poor
minors
and
I
thought
hey.
A
This
might
be
my
first
chance
at
a
customer,
so
david
emailed
me
with
some
feedback.
He
emailed
me
and
said
he
loved
he
loved
to
slack
integration,
but
he
was
wondering
if
it
could
not
send
reminders
on
weekends
because
it
would
disturb
his
team.
So
I
thought
to
myself
easy
I'll
just
add
a
mute
on
weekends
checkbox.
A
Well,
he
got
back
to
me
and
said
that
actually,
his
weekends
were
Friday
to
Saturday
and
not
Saturday
Sunday,
because
his
team
is
based
in
Israel,
so
I
thought:
okay,
yeah
I
got
this
I'll
just
I'll
just
add
a
checkbox
for
every
day.
So
you
can
pick
exactly
the
days
you
want
reminders,
and
so
you
know
I
was
desperate
to
get
this
first
customer.
So
I
did
this
right
away.
A
It's
two
two
full
days
and
got
got
it
back
to
him
and
he
got
back
to
me
and
said
it
was
working
great
but
that
he
had
one
more
feature:
request.
Well,
I
cut
this
email
short.
It
was
actually
like
five
feature
requests
and
while
I
was
I
was
really
excited
to
go
back
and
forth
with
David.
You
know
get
his
feedback
on
my
product,
I
also
started
to
get
anxious.
Yeah
I
was
worried
that
I
was
doing
all
this
for
free
and
I
wanted
to
get
paid.
A
So
I
decided
to
ask
the
question:
I
popped
the
question
and
asked
David
if
he'd
be
willing
to
pay
fifteen
dollars
a
month
for
full
reminders,
and
this
is
just
a
completely
arbitrary
number.
It
seemed
you
know
high
enough
to
be
meaningful
but
low
enough
that
you
probably
won't
say
no,
and
he
got
back
to
me
and
said
that
they
were
definitely
willing
to
pay
for
something
that
gave
them
value
and
I.
A
Remember
being
really
confused
by
this,
because
I
wasn't
sure
if
you
met
yeah,
you
know,
poll
reminders
was
giving
him
value
or
no,
it
wasn't
giving
him
enough
value
and
he
went
pack.
Anyways
David
said
that
you're
still
testing
it
out
and
that
he
was
gonna
actually
show
it
to
management
once
he
was
ready
and
so
yeah.
A
Whatever
we
kept
going
back
and
forth
for
a
couple
weeks
back
and
forth
back
and
forth,
you
would
send
me
feature
requests
and
I
would
still
do
em
and
you
know
I
still
hadn't
gotten
paid,
so
I
got
really
anxious
and
a
couple
weeks
later,
I
popped
the
question
again
and
this
time
I
asked
him.
If
he'd
be
willing
to
pay
twenty
dollars
a
month
to
use
poll
reminders
I,
don't
know
why
I
raised
the
price
five
dollars,
I
think
I
was
just
feeling
resentful
and
why
he
got
paid
anyways.
A
My
advice
is
make
sure
you
talk
to
your
early
users,
and
some
people
might
disagree
with
this,
but
I
think
you
should
pretty
much
do
whatever
they
asked
for.
Even
if
that
takes
you
in
unexpected
directions
or
know
things
that
you
don't
like
chances
are.
What
they're
gonna
do
is
take
your
half-baked
idea
and
turn
it
into
something
that
really
delivers
value.
A
So
I'd
gotten
a
couple
customers
and
I
figured
I
was
gonna,
get
some
more
hopefully,
and
so
I
decided
I
needed
to
come
up
with
some
real
pricing
instead
of
just
quoting
people.
You
know
arbitrary
numbers
based
on
how
I
was
feeling
and
so
I
knew
that
poor
reminders
was
gonna,
be
price
based
on
the
number
of
users
on
the
team,
but
I
didn't
know
how
much
to
charge
and
yeah
I
was
not
feeling
so
confident
at
this
point
still
and
wanted.
You
know
wanted
something
that
would
be
hard
to
say
no
to
so.
A
I
asked
myself.
What's
something
else,
that
developers
and
engineering
teams
are
willing
to
pay
for
for
productivity
and
what
came
to
mind
was
coffee,
so
I
spent
a
lot
of
time.
You
know,
thinking
about
you,
know,
positioning
whole
reminders
against
coffee,
I,
even
thought
about
putting
the
slogan
on
my
website
productivity
for
less
than
a
cup
of
coffee.
I
guess
it
depends.
You
know
where
you're
buying
your
coffee.
This
is
this
was
based
on
the
Chicago
price
of
coffee,
so,
based
on
that
I
came
up
with
a
rate
of
$2
per
user
per
month.
A
This
is
my
pricing,
so
you'll
see
that
actually
I
don't
do
a
flat
rate
and
instead
I've
pricing
plans
and
the
reason
I
do
that
is
so
I
can
make
more
money.
So,
for
example,
if
a
team
is
nine
developers
instead
of
paying
me
a
flat
rate
of
nine
times
to
$18
a
month,
they
have
to
upgrade
to
the
25
developer
a
month
plan.
So
I
make
49
a
month,
and
you
know
this
is
actually
tricky.
A
Sometimes
I
actually
get
users
who
email
me
and
they're
on
the
fence,
because
that
jump
from
14
to
49
seems
steep.
So
you
know
once
in
a
while
I'll
give
someone
the
discount
and
give
them
sort
of
an
in-between
plan.
You'll
also
see
that
the
75
developer
a
month
plan
is
not
$2
per
user
per
month.
It's
actually
less
and
again.
I.
Don't
have
some
fancy
reason
for
this.
It's
just
that
I
had
trouble
getting
people
to
pay
more
and
so
I
decided
to
settle
a
little
bit
and
bring
that
price
down.
A
I
give
pull
reminders
away
for
free
to
open
source
projects,
but
I
want
to
highlight
the
terms
non-commercial.
When
I
first
launched
Laura
miners
I
started
getting
these
big
open
source
based
companies,
companies
like
Cloud,
Foundry
and
Sun
Tzu,
and
they
were
signing
up
for
poor
reminders
for
free
and
so
I
decided
to
have
the
term
non-commercial
so
that
poor
reminders
would
only
be
free
for
open-source
projects
and
not
big
tech
companies
that
have
an
open-source
business
model
in
general.
I
think
pricing
is
difficult.
A
A
A
Emailing
me
talking
about
the
same
problems,
but
they
had
different
ideas
for
how
I
should
solve
it,
and
my
brain
would
just
kind
of
explode
because
I
wouldn't
know
you
know
how
to
reconcile
these
different
ideas
and
pick
the
best
solution
forward.
So
I
tried
this.
What
I
did
is
I
created
this
public
backlog
using
github
projects,
and
so
this
is
public.
A
If
you
go
to
the
poll
reminders
or
you
can
check
it
out,
submit
issues
and
I
designed
it
like
a
pretty
traditional
Kanban
board,
so
it
has
to
do
you
know
in
progress
and
done
and
when
I
created
this
I
wasn't
sure
was
gonna
happen,
so
I
just
started
using
it
to
track
my
own
work,
so
whatever
I
was
working
on,
I
would
put
in
this
backlog,
and
you
know
I
would
share
it
with
customers
over
email.
Today,
this
board
has
around
80
issues,
I
think
and
every
one
to
two
weeks.
A
I
get
a
user
submitted
issue,
and
this
is
what
that
looks
like.
So
you
guys
are
all
familiar
with
this,
but
yeah
a
customer
will
submit
an
issue
and
then
I'll
tag.
It
I
then
prioritize
it
in
the
backlog
and
like
I
mentioned
it's,
not
just
user
submitted
issues.
I
often
create
my
own
issues
based
on
the
conversations
I'm
having
with
users
over
email
or
you
know,
support,
and
that
one
challenge
I
mentioned
earlier,
where
you
know
different
users
would
give
me
opinions
on
the
same
things,
but
with
different
solutions.
A
This
is
one
thing
that
having
an
open
backlog
has
really
helped
with.
So
when
I,
a
suggestion
from
a
customer
or
if
I
have
my
own
idea,
I'll,
actually
ping
a
lot
of
a
lot
of
my
users
on
github
to
get
their
input
and
yeah.
This
is
an
example
of
that,
and
so
what's
interesting
about
this
I
think
is
it's
sort
of
like
a
open
source
model
for
product
development.
Even
though
poll
reminders
it's
closed
source-
and
this
has
ended
up
working
really
well
for
tracking
ideas,
sharing,
updates
and
helping
me
get
input
from
users.
A
So
I've
gotten
a
couple:
customers
at
this
point
and
I
was
really
eager
to
get
more
I
wanted
to
quit.
My
contractor
work
and
go
full-time
on
pull
reminders
and
you
know
a
lot
of
startup
frameworks
out
there.
They
focus
on
you
know,
coming
up
with
business
ideas
and
validating
your
problem
and
things
in
that
nature.
About
for
me
and
for
a
lot
of
other
independent
entrepreneurs.
I
talked
to
the
hardest
part
of
building
a
business
is
actually
reaching
your
target
audience.
A
So
you
can
make
money
now
you
can
have
the
best
product
in
the
world
with
great
messaging,
but
if
you
can't
actually
get
in
front
of
people,
you're
not
gonna
make
any
money,
and
so,
while
I
was
working
on
building
poor
reminders,
I
was
reading
the
github
documentation
and
I
heard
about
this
thing.
Called
github
marketplace
raise
your
hand
if
you've
heard
of
github
marketplace.
A
Okay,
this
is
a
github
conference
for
me,
I
had
been
working
with
github
for
seven
years
and
I
never
heard
of
github
marketplace,
and
so,
when
I
saw
this
I
was
like
amazing.
You
know
this
could
solve
my
problem
of
reaching
my
target
audience
so
for
those
of
you
who
aren't
familiar
with
github
marketplace,
there's
actually
a
link
in
the
very
main
menu
of
github.com
called
marketplace
and
github
marketplace
is
a
directory
of
third-party
apps
and
github.
A
Approves
these
apps
and
once
an
app
is
listed,
people
are
able
to
purchase
them
through
github
in
exchange
for
github,
taking
a
25%
cut
of
your
revenue,
and
so
I
thought.
This
looks
like
a
great
opportunity
and
I
decided
to
apply
and
I.
Remember
when
I
applied,
I
was
super
nervous.
You
know,
I
didn't
you
know.
Poor
miners
was
still
kind
of
this
sort
of
a
side-project
couple.
Customers
and
the
first
step
in
the
process
was
I
got
an
email
from
a
guy
named
Steve
who
works
at
github.
He's
a
partner
engineer.
I.
A
Remember
when
I
got
this
email
I
was
I
was
terrified
felt
like
a
job
interview.
You
know,
I
worried
what
if
Steve
was
mean
what,
if
he
didn't
like
poll
reminders
what,
if
he
told
me
I,
would
never
be
allowed
in
the
marketplace
anyways.
This
is
Steve
and
when
we
hopped
on
our
call
my
fears
on
the
way
Steve
was
super
nice
relaxed
about
the
whole
process,
charming
English
accent,
and
you
know
he
outlined
the
process
for
me
and
over
the
course
of
the
next
couple
of
weeks.
A
This
is
poll,
reminders,
went,
live
on
the
github
marketplace,
April
3rd,
and
this
is
what
the
listing
looks
like
and
what
it
looked
like.
Still
the
same,
and
once
again,
I
tried
to
launch
with
the
tweet
again
secretly
hoping
that
you
know
we
got
tons
of
retweets
and
go
viral,
and
that
didn't
happen
again.
This
time.
I
did
make
some
money,
though
I
think,
and
the
best
part
of
launching
on
the
github
marketplace.
Wasn't
the
fact
that
I
actually
got
listed.
A
It
was
the
support
I
got
from
people
who
worked
at
github,
so
all
these
people
are
github
employees
and
Kirk.
I'll
speak
and
I'm
going
to
tell
a
little
story
about
Brandon,
but
he's
actually
the
guy
who
built
the
official
github
slack
integration-
and
you
know
that's
just
some
of
the
people
I
got
support
from
these
are
a
bunch
of
other
people
at
github
that
I've
personally
interacted
with
and
have
supported
me
along
the
way
and
see
I
work
alone,
and
you
know
sometimes
working
on
poor
reminders
feels
kind
of
small.
A
You
know
it's
just
kind
of
a
silly
app
and
so
having
these
interactions
with
people
like
github
and
getting
their
support
has
made.
It
made
me
feel
like
I'm
part
of
github
mission,
and
this
makes
me
feel
like
I'm
working
on
something
larger
than
myself,
and
you
know
that's
been
really
meaningful,
so
funny
story
about
Brandon
who's
here
in
the
audience
today
about
a
week
before
I
went
live
on
the
github
marketplace.
I
woke
up
one
morning,
went
on
Twitter
and
heard
that
github
had
just
launched
their
own
new
integration.
For
slack
and
I.
A
Remember
my
heart,
just
sank:
yeah
I
was
I,
figured
I
was
toast,
I
figured
that
get
of
it
just,
but
the
exact
same
thing
as
me
and
I
just
wasted
all
my
all.
My
time
and
pull
reminders
would
be
irrelevant,
so
I
went
online
right
away
and
all
of
it,
and
at
least
for
the
time
being
it
looked
like
I,
was
safe.
There
wasn't
any
overlap
in
the
features
and
no
so
I
was
reading
about
this
new
integration.
A
It
launched
on
product
hunt
and
got
tons
of
up
votes
and
people
on
product
hunt
kept
talking
about
this
guy
Brandon.
They
said
you
know
Brendon
this
guy's,
awesome
and
beekeepers,
and
so
I
was
like
who's
Brandon.
So
this
is
Brandon
and
turned
up.
Brandon
lived
in
Chicago,
where
I
also
was
at
the
time
and
I
decided
to
reach
out
I
figured.
Maybe
if
Brandon
liked
me,
he
went
crushed
pull
reminders.
Who'd
spare
me
from
github
Strath,
and
you
know,
if
nothing
else,
I
figured.
A
He
could
give
me
some
advice
on
how
to
be
successful
on
the
github
marketplace.
Again,
I
was
really
nervous
to
me.
Brian
Brandon
I,
you
know,
felt
like
a
job
interview.
Well,
Brandon
was
also
super.
Nice
super
warm
and
after
talking
for
a
while
after
a
couple
drinks,
he
actually
asked
me
directly
if
I'd
been
scared.
When
I
had
heard
about
his
new
integration
and
I
said
yes,
I
was
scared.
He
laughed
at
me
inside.
A
You
have
nothing
to
worry
about,
there's
nothing
on
the
roadmap
like
poor
reminders,
and
this
is
a
really
special
moment
for
me.
You
know,
I
think
whenever
you
build
something
on
the
third
party
platform,
you
have
legitimate
fears
that
that
platform,
might
you
know,
kick
you
off
or
directly
start
competing
with
you
and
so
for
me,
all
these
interactions
with
people
at
github
have
made
it
clear
to
me
that
github
has
really
invested
in
their
ecosystem
strategy,
and
that
least
for
the
time
being,
my
business
is
secure.
A
A
July
is
when
I
was
able
to
cover
my
bills
and
go
full
time
on
pool
reminders.
So
this
is
obviously
a
big
milestone
for
me
and
on
this
chart,
I
don't
just
show
paid
accounts,
I
actually
show
the
total
number
of
installations
and,
like
I
mentioned
earlier,
that's
because
I
give
away
pull
reminders
for
open
source
projects
and
also
some
nonprofits
and
student
organizations.
A
So
I
work
alone
by
myself,
which
is
a
bit
weird,
sometimes,
but
overall
I
love
it
and
the
reason
I
love
it
is.
You
know
one
of
the
main
reasons
that
I
wanted
to
start.
My
own
business
was
to
have
a
lot
of
freedom
and
in
the
past,
I've
tried
building.
You
know,
building
projects
of
people
co-founders
and
what
I
found
that
what
I
found
was
that,
even
though
I
was
technically
working
for
myself,
I
didn't
feel
free.
You
know
my
co-founder
would
be
like
hey.
You
know.
A
If
I
wanted
to
take
a
day
off,
my
co-founder
would
be
like
hey,
we
got
a
hustle,
we
got
to
work,
and
so
I
won't
be
able
to
take
a
day
off
so
working
by
myself.
Now,
I
truly
am
my
own
boss
and
I
love
it
because
you
know
I
can
go
about
my
work.
The
way
I
want
and
I
have
complete
creative
autonomy
over
my
product.
This
is
a
little
quote
by
guy
Derek
Siver's.
A
If
you
haven't
heard
of
him,
he's
a
programmer
sold
a
company
to
Apple
in
the
90s,
and
this
is
actually
in
his
bio.
So
you
know
he
talks
about
working
alone
and
it
really
resonates
with
me
for
me,
working
on
poor
reminders
is
this
very
personal,
creative
pursuit,
so
working
alone
is
pretty
fun,
but
one
of
the
downsides
is
I
have
to
do
everything
and
including
things
that
I'm
not
comfortable
with
and
things
I
don't
want
to
deal
with,
and
so
one
of
those
challenges
for
me
has
been
figuring
out
how
to
support
github
enterprise.
A
So
we
are
at
a
github
conference,
but
if
you're
not
familiar
with
github
enterprise,
its
github
self
hosted
version
of
github
that
usually
large
companies
are
able
to
download
and
run
on
their
own
infrastructure
within
their
corporate
networks.
So
for
me
you
know
this
is
something
I
put
off
for
a
long
time.
It
seemed
really
hard
and
I
wasn't
good
at
this
type
of
stuff
and
a
few
months
ago,
a
guy
actually
who
worked
at
PayPal,
had
reached
out
to
me
and
said:
hey,
let's,
pull
reminders
work
with
get
a
better
prize.
A
I
told
him,
no,
not
yet
maybe
soon,
but
he
kept
following
up
with
me
every
three
weeks
asking.
Is
it
done
yet?
Was
it
done
yet?
Was
it
done
yet
so
eventually,
I
decided
I
had
to
give
in,
and
the
process
of
delivering
I'm
premise?
Is
software
I'd
never
done
this,
so
it
was
completely
new
to
me
it's
very
different
than
delivering
a
hosted
piece
of
software.
You
know
I
had
to
rethink
everything
from
you
know
how
to
track
errors
on
the
customers
infrastructure.
A
A
Anything
DevOps
has
ever
come
up
in
my
career,
I've
just
hit
or
run
the
opposite
direction,
and
probably
before
this,
the
hardest
thing
I'd
ever
done
was
set
up
a
docker
file
and
95%
of
that
was
just
copying
and
pasting.
So
long
story
short
the
solution
I
came
up
with
was
these
docker
and
I
chose
docker
because
it
seemed
like
the
lowest
common
denominator.
A
You
know
something
that
could
work
regardless
of
the
type
of
host
or
machine
that
a
customer
was
interested
in
the
only
thing
I
did,
which
some
of
you
might
be
offended
by
was
I,
stuffed
my
entire
application,
except
for
Postgres
and
one
docker
image.
So
you
know
poor
reminders,
it's
a
rails
app,
but
it
has
a
worker
process
dependencies
like
Redis-
and
you
know,
I
wanted
to
avoid
dealing
with
anything
like
kubernetes
or
anything
like
that.
So
another
tricky
problem
was
licensing.
A
A
Aka
lazy
and
I
didn't
want
to
deal
with
this,
and
so
what
I
ended
up
doing
was
really
not
doing
anything
at
all
for
licensing
and
today,
I
just
agree
on
licensing
in
a
contract
with
the
customer
and
it's
an
honor
system
where
I
just
expect
them
to
adhere
to
it.
I
also
made
some
pretty
embarrassing
mistakes
figuring
out
how
to
support
github
enterprise,
mostly
out
of
efficiency,
aka
laziness.
One
of
the
mistakes
I
made
is
I,
never
actually
downloaded
and
ran
github
Enterprise
myself
to
test
against
that.
A
You
know
ironed
out
all
the
kinks
and
one
of
the
problems
as
a
result
of
this
was
that
I
didn't
realize
that
there
were
multiple
versions
of
github
enterprise
and
it
turns
out
that
these
different
versions
are
very
well
pretty
different
api's,
and
so,
when
I
was
getting
my
first
customer
trying
to
get
him
set
up.
We
kept
running
into
these
errors,
I'd
never
seen
before
and
turned
out.
A
A
A
So
another
challenge
of
working
by
myself
is
staying.
Positive,
I,
don't
have
you
know
a
partner
or
someone
to
emotionally
support
me
when
things
get
difficult
or
I
have
to
deal
with
things
like
github
Enterprise,
and
so
for
me.
You
know
if
someone
signs
up
for
a
free
trial,
full
reminders
and
then
decides
not
to
purchase
a
subscription.
You
know
things
like
back
and
sort
of
sent
me
in
the
dark
funk,
and
you
know
I
get
down
for
a
couple
of
days
today.
A
A
One
last
funny
story
about
working
alone,
so
at
the
beginning
of
the
year,
I'd
planned
on
going
on
a
backpacking
trip
with
my
dad
and
brother
in
the
Grand
Tetons.
So
this
is
going
to
be
a
week-long
backpacking
trip
with
no
internet.
No
cell
signal
and
close.
This
is
last
month
closer
and
closer
to
the
trip.
A
So
I
looked
in
the
how
I
could
try
to
deal
with
this
and
I
I
looked
up
the
cell
coverage
maps
for
the
Grand
Tetons,
and
you
know
that
wasn't
gonna
work,
I
even
considered
actually
lugging.
My
backpack
with
me,
while
backpacking
and
I
thought
about
trying
to
hire
someone,
but
I
didn't
really
have
the
money.
A
For
you
know
people
like
me:
if
we
need
to
take
a
vacation,
and
my
advice
would
be
you
know
if
you're
backpacking,
somewhere
with
Brizzy
Bears,
and
if
you
see
a
bear
that
doesn't
see
you
don't
start
screaming
like
I,
did
won't
be
good,
so
I
hope
you've
enjoyed
my
story
and
that
it's
been
helpful
and
I
hope
you've
seen
that
building
a
business
is
actually
a
lot
like
building
software.
If
you
break
it
up
into
these
small
pieces,
it's
quite
manageable
and
actually
fun
so
I'm
sure.