►
From YouTube: 2023-09-19 Product Analytics Group Sync
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Cool
hello,
everyone
and
welcome
to
the
September
19th
how
to
really
quickly
check
there,
because
I
forgot
product
analytics
group
seek.
We
can
go
ahead
and
get
started
with
the
billboard.
Review.
I
will
share
my
screen
and
then
we
can
do
it
cool.
So
a
number
of
issues
in
open
surrounding
cost
data.
The
usage
limit
we'll
talk
more
on
that
later
164
updating,
Cube
schemas
to
exclude
Anonymous
users,
I
got
yeah,
I,
don't
think
that's
started
yet.
We've
just
started
the
migration
of
those,
so
I
think.
A
Unless
anyone
believes
we
can
discuss
any
of
these,
we
can
just
move
on
to
those
with
workflow
labels.
A
All
right,
cool
first
issue
is
blocked,
so
we
can
move
on
there.
We
have
handling
on
Save
changes
analytics
dashboard
Eon.
Are
you
able
to
speak
to
this
right
now?
Aren't
you.
B
Yeah
I
can
speak
to
this
quickly.
It's
in
progress
It's
turned
out
to
be
a
slightly
larger
change
than
I,
initially
thought,
because
not
only
it's
not
as
simple
as
just
resetting
the
dashboard
to
the
initial
object
and
like
dropping
the
graphed
object,
because
grid
stack
still
retains
all
the
panels.
In
the
background,
so
I'm
now
working
to
figure
out
what
the
exact
Delta
is
or
finding
a
way
of
resetting
grid
stack.
It's
a
fun
sign,
but.
A
B
I
might
need
to
set
up
a
pairing
session
with
rope
if
my
brain
cannot
figure
this
out
by
the
end
of
today.
So
that's
good.
A
Awesome
and
next
issue
is
adding
product
analytics
to
the
experiment,
feature
list.
Next,
you
have
this
one
assigned
to
you
anything
you
can
speak
to
here.
D
Yep
so
this
was
part
two
of
a
two-part
issue,
first
of
which
was
to
move.
The
experiment,
features
for
a
namespace
out
of
sort
of
an
AI
coupled
environment
which
is
done
in
production,
and
the
next
is
to
lay
the
groundwork
to
allow
anyone
to
opt
in
to
product
analytics
at
the
moment.
I'll
be
merging
them
out.
D
That
puts
it
behind
that
flag,
but
also
behind
a
million
other
feature
flags
as
well,
so
there'll
be
no
functional
difference,
except
that
existing
customers
will
need
to
turn
on
experimental
features
on
their
namespace
gitlab
org
already
has
it
turned
on,
so
there
won't
be
any
effect
on
that
one.
But
if
we've
onboarded
an
external
customer,
they
will
need
to
enable
it
if
they
have
not
already
done
so.
They
can
do
that
today
if
they
want
to.
A
Okay,
I'm
not
familiar
with
the
process
to
do
so,
but
we
may
want
to
forward
that
to
our
external
customer,
which
we've
just
enabled
this
week,
could
you
share
information
with
James.
D
Yeah
sure
absolutely
I
mean
the
quick
version
is
in
the
namespace
settings.
There's
a
experimental
features
toggle
that
they
just
need
to
toggle
on
to
an
owner
in
their
namespace
needs
to
do
that
got
it.
A
Awesome
I
have
the
next
one
for
customer
facing
infrastructure
release,
so
we've
gotten
a
lot
of
what
we
want
done
here
and
I
just
clicked
the
link
without
opening
any
tabs,
that's
cool,
so
we've
made
a
lot
of
progress
here,
namely
we've
got
all
the
security
review
approvals
that
we
need
to
really
launch
this.
A
At
the
same
time,
we're
getting
closer
and
closer
to
getting
out
to
the
infrastructure,
managed
environments.
Pierre
has
everything
set
up
too.
This
is
confident,
but
there's
something
detailed
here
that
really
is
leaking
anything
other
than
just
the
fact
that
it
handles
red
data
which,
if
anything,
is
less
of
a
severity
given
that
we
don't
host
for
a
data
anymore,
but
we
continue
to
make
progress.
Pierre
has
the
pre-environment
set
up,
which
is
supposed
to
be
kind
of
a
reference
architecture
to
staging
and
production.
A
I
guess
this
before
we,
it
even
gets
to
staging
actually
so
we're
going
to
verify
that
everything
works
there
before
we
move
on
to
staging
and
then
eventually
production,
so
we're
getting
closer
and
closer.
A
At
the
same
time,
we
need
to
figure
out
what
run
books
we'll
need,
but
I
think
that's
really
going
to
be
an
event
trying
to
like
really
get
all
the
infrastructure
topics,
kind
of
collected
and
organized,
but
that's
really
going
to
be
dependent
on
logging
and
monitoring
and
understanding
how
to
scale
things
or
react
to
Things,
based
on
what
part
of
the
stack
is
actually
failing
or
having
issues
so
I'm,
not
sure,
really
what
we
can
really
do
now,
but
basically
the
more
I
think
about
it.
A
The
more
that
I
can
speak
to
at
a
later
point,
but
all
that
to
say
is:
we've
made
good
progress
here,
I'm
trying
to
figure
out
how
we
can
kind
of
close
this,
because
we've
really
already
finished
conducting
the
security
review
and
we're
serving
production,
albeit
the
last
step,
is
really
to
get
into
the
infrastructure,
management
and
production
environment.
A
So
if
anything
I've
just
answered
my
own
question
by
explaining
it
that'll
be
the
last
step.
So,
thanks
for
listening
to
my
TED
Talk
cool
I
will
go
back
to
the
billboard.
A
And
we
can
continue
down
the
list,
so
we've
got
updating
the
project,
Technologies
cluster
settings
field,
so
we've
actually
got
the
next
two
issues.
Do
you
want
to
speak
to
any
of
these.
E
Oh
yeah
sure
yeah
they're,
currently
in
development,
the
second
one
I'll
put
for
a
review.
Today,
it's
almost
done
yep.
A
C
Yeah
sorry
I
just
hadn't
done
that
to
myself.
Although
I
did
this
on
the
Mr
just
assigned
myself,
both
of
those
are
currently
blocked,
yarn
just
merged
the
gitlab.com
changes
to
enable
the
newly
named
schemas
on.com
as
soon
as
that's
verified
in
reduction.
These
are
no
longer
blocked.
I've
already
got
Mrs
ready
to
go
for
both
the
dev
kit
and
analytics
stack
for
the
Final
Phase
of
working,
and
this
can
be
closed
off.
A
Good
awesome
and
then
we've
got
actually
Four
issues
owned
by
Elwyn
I'm,
just
gonna
quickly
check
to
see
if
there's
any
update
here,
nope
and
then
improving
dark
mode
error,
States,
the
visualization
selector
and
a
bug
with
still
in
progress
on
a
give
audio
IMR
in
our
states.
The
visualizations
in
review
and
there's
also
a
review
for
editing
a
dashboard
duplicates
on
the
dashboard
list
and
changes
made
are
lost
so
just
moved
into
review
yesterday.
So
good
work.
E
C
Yeah,
this
is
still
with
Cube
I've,
not
heard
a
word
from
them,
although
one
of
their
team
has
assigned
it
to
themselves,
but
nothing's
come
about.
So
if
I
don't
hear
anything
by
the
end
of
the
week,
I'll
start
nudging
next
week
to
see
if
I
could
get
some
kind
of
traction,
I'm
asking
the
slack
Channel
as
well
awesome.
A
Thanks
for
pushing
for
Action
there,
cool
I've
got
an
issue
from
Alan
who's
no
longer
on
the
team,
but
he
just
doesn't
want
to
make
sure
that
this
gets
wrapped
up.
So
I
believe
he's
still
looking
to
verify
this
to
make
sure
that
whatever
service
paying
metrics
we've
set
up
at
this
point
are
starting
to
show
up
and
then
rob
you
have
the
next
two
issues
in
verification.
C
Yeah,
the
first
ones
I
mentioned,
is
to
do
like
git
lab
repo
change
to
start
using
new
schemas
I'm.
Just
waiting
for
that
to
hit
staging
Etc,
so
I
can
start
verifying
it,
and
the
line
chart
was
a
bug
that
Sam
found
I.
Believe
yes,
Sam.
E
C
Yes,
where
we
weren't
defining
the
type
on
the
y-axis,
which
is
throwing
errors
on
the
visualization
designer,
so
just
fixing
that
small
little
bug
and
waiting
for
it
to
deploy
so
I
can
test
it.
A
Awesome
cool:
we
don't
have
a
book
Mark
for
the
past
three
six,
so
I'm
just
going
to.
A
I
want
to
say
I
think
perhaps
renaming
the
graphql
type
slider
than
where
we
kind
of
stopped,
but
maybe
someone
correct
me
well,
okay,
let
me
just
do
it
this
way.
I
can
call
it
a
couple
of
issues
that
we've
closed
recently.
That
I
think
we
can
highlight
We've
connected
clickhouse
Cloud
to
the
analytics
stack.
A
That's
been
good.
We
have
that.com
going
as
well
as.
A
Ongoing
project
in
production,
as
well
as
an
external
customer,
so
it'd
be
exciting
to
see
how
that
goes
related
once
we're
actually
in
the
infrastructure
manager
environment.
That
will
also
be
pointed
to
the
same
clickhouse
instance,
so
that
should
all
continue
to
work,
but
we
may
have
to
change
some
instant
settings,
but
hopefully
I'm
working
with
Pierre
to
come
up
with
a
migration
plan,
but
hopefully
should
have
minimal
disruption,
especially
with
the
external
customer.
At
least
the
data
shouldn't
be
going
anywhere,
different
I
think
that
would
also
cover
yeah.
C
A
C
Completed
one
thing,
I'd
call
out
is
actually
one
of
elwin's
further
down
use
new
qgs
cost
numeric
option,
just
as
a
highlight
the
renovate
bot
pinged
me
now
win
on
a
new
release
of
cube
GIS
and
when
we
were
reviewing
the
changelog,
we
noticed
that
they've
released
a
new
cast,
numerix
parameter,
which
oh
and
then
immediately
went
ahead
to
implement
within
our
code
base,
and
it
basically
now
means
that,
rather
than
returning
strings
for
every
value,
as
you
return
to
the
right
type
for
the
data,
which
is
far
easier
to
handle
and
much
nicer
to
you,
so
just
a
shout
out
for
renovate
bot,
and
also
just
for
L
when
tackling
it
so
quickly
and
getting
it
into
our
code
base.
A
That's
awesome:
yeah
I
can
see
that
now
it's
showing
in
actual
numbers
and
prop
sale,
one
for
the
bias
for
Action
there
and
I'm
glad
we
have
another
bot,
so
I
need
to
pay
attention
to
the
new
releases,
so
also
great
I
will
continue
to
scroll
for
about
10
more
seconds.
If
anyone
would
like
to
call
anything
out.
Otherwise
we
can
wrap
up
the
Board
review.
A
All
right,
then
I
will
set
a
bookmark,
so
we
don't
forget
and
we
can
continue
with
the
rest
of
the
agenda.
D
D
There's
been
a
few
conversations
about
it,
so
I'm,
currently
working
I,
say
I'm
currently
working
me,
Rob
and
halil
are
currently
working
on
an
MR
for
the
configurator
to
pull
out
per
project
the
number
of
events
being
stored
on
a
monthly
basis
that
is
taking
longer
than
expected,
because
I
am
not
an
accomplished.
Golang
developer,
but
I'm
learning
a
lot.
So
it's
been
interesting.
So
thank
you
for
the
two
of
you
for
helping
and
Nico
as
well.
D
D
So
the
the
aim
is
that
we'll
create
a
graphql
interface
so
that
the
front-end
Engineers
can
pull
out
per
month.
How
many
events
a
project
is
using
and,
by
extension,
how
many
events
the
namespaces
projects
we're
using
I've
defined
in
the
linked
issue.
There
a
couple
of
potential
graphql
queries
that
could
be
used
a
few
ideas,
so
yeah
I
wanted
to
open
it
up
to
what
folks
I'm
thinking.
D
There's
a
couple
of
assumptions
I've
made
as
well
about
how
we're
counting
used
data
and
the
the
dates
that
we
use
for
that
yeah.
That's
it
I
just
wanted
to
open
it
up
really.
D
Please
review
it.
I
know,
I've.
Just
seen
that
Rob
sort
of
responded
to
the
bigger
question
that
was
open
and
then
I
see
James
has
already
responded.
Actually,
the
two
big
questions
I
had
were:
are
we
counting
by
data
or
by
events
stored
and
are
we
able
to
assume
that
we
will
be
billing
on
a
calendar
month
basis,
as
opposed
to
not.
F
Yes,
the
idea
is
that
we're
going
to
be
billing
based
on
some
sort
of
event,
count
and
probably
based
on
the
calendar,
we're
still
working
at
a
lot
of
the
logistics
of
that
with
oh,
my
group,
but
for
now
that
aligns
with
how
we
do
CI
our
compute
credits.
So
it
makes
sense
to
follow
that
same
pattern
and
I'll
have
some
other
notes
around
flip-flopping
the
iterations,
to
make
it
easier
for
the
MVC.
Really
that's
how?
How
many
events
are
you
spending?
F
Ideally,
we
can
see
it
by
project,
but
if
we
want
to
start
just
at
the
group
level
and
then
kind
of
force,
a
click
and
wait
to
get
two
projects
for
the
MVC,
that
would
be
an
okay
user
interaction
and
then
we
can
iterate
from
there.
D
To
be
honest,
the
project
is
easier
to
count
than
than
groups,
because
each
project
contains
its
own
data.
So
we
just
say
it's
a
h
table.
Sorry.
So
if
you
just
say
give
us
the
the
count
of
events
in
that
table,
yeah
namespace
is
potentially
more
complex
because
we're
adding
n
number
of
projects
together.
So
if
a
particular
namespace,
for
example,
had
a
thousand
projects
using
products
analytics
which
is
not
super
likely.
F
Yeah
I
mean
take
a
look
at
the
the
mock,
that's
there
and
let's
start
with
the
easiest,
visualization
or
facet
visualization
I'll,
leave
it
to
y'all
to
figure
out
which
Opus
pal
can
take,
and
we'll
start
with
that
in
the
MP3,
and
we
can
iterate
into
a
better
user
experience
or
a
different
graphical
experiences.
If
you
need
to
I,
think
table
is
probably
the
best
there
versus
graph
okay.
C
Whatever
visualization
doesn't
particularly
matter
once,
we've
got
the
data
from
a
front-end
point
of
view.
We
just
plan
that
data
into
that
visualization
and
produce
the
output
so
either
visualization
doesn't
make
a
difference.
We
just
need
to
make
sure
that
we
get
all
the
projects
data.
Essentially,
ideally
in
one
go,
we
tried
to
do
it,
we're
gonna
if
we
tried
to
do
per
project
or
in
the
sense,
if
we
have
to
make
single
individual
API
calls
for
each
project.
C
That
would
be
a
big
bad
I'd,
be
worse
than
groups
when
it
comes
to
exponential
performance
issues.
So
if
we
can
just
make
one
API
call
which
batches
all
the
projects
of
a
group
and
gives
us
that
data
breach
project,
then
we
can
handle
it.
If
we
need
to
paginate
happy
to
work
out
how
to
paginate
table
would
be
the
simplest
solution
for
that,
because
we'd
need
to
be
a
bit
longer
to
try
and
figure
out
how
to
impassionate
a
graph.
D
Are
you
happy
if
you
only
had
to
make
one
graphql
query,
but
the
response
was
an
array
of
projects,
so
you
know
you
have
yes,
a
potentially
10
to
15
numbers
to
add
up.
That
would
be
fine.
C
Well,
we
don't
need
to
add
them
up.
We
just
need
to
plumb
all
the
so
we
just
the
projects
returned.
We
just
be
a
with
the
project
with
its
usage
over
that
time
period
and
we
just
plumb
that
data
in
to
the
graph.
So
we
don't.
The
only
time
we
need
to
add
up
is
when
we
go
to
the
group
level,
but
that's
the
second
iteration
we've
discussed
and
we're
not
even
considering
that
just
yet
so
yeah
having
a
big
array
of
group
or
projects
is
absolutely
fine.
D
A
A
Any
other
questions
refer
to
the
issue.
Thanks
thanks,
Max
James
had
to
read
only
point
I,
don't
know
if
you
want
to
verbalize
that
or
if
we
just
move
on.
F
Yeah
great
work,
everyone
Kudos!
Thank
you.
You
two
huge
amounts
of
this
week.
Getlive.Com
has
project
analytics,
AdWords
I,
know
that
was
a
big
list
of
infrastructure
and
just
all
of
the
work
over
the
last
year
and
first
external
customer
on
board
and
still
waiting
on
events.
They've
got
a
little
radio
asylum
on
us,
but
I'll
keep
poking
and
prodding
and
get
some
feedback
from
the
customer
on
how
things
are
working.
F
I'll
be
sending
a
note
out
to
what's
happening
in
gitlab
with
some
of
the
details
there
and
so
expect
to
get
some
feedback
in
the
analytics
section
Channel
over
the
course
of
the
day.
A
Exciting
stuff
awesome,
so
165
planning
issue
has
a
bit
more
substance.
Now
it's
now,
if
you've
been
looking
at
the
board
view,
it
hasn't
really
particularly
changed,
but
to
kind
of
help
keep
everything
collected
in
the
planning
issue.
It's
gone
ahead
and
added
everything
there,
as
well
as
a
couple
other
topics
that
we
were
looking
at.
A
So
at
this
point
the
planning
issue
and
the
billboard
should
be
in
sync-
is
what
I'm
trying
to
say
but
I
think
James.
If
you're
still
good
with
it,
we
can
go
over
the
kind
of
themes
for
the
Milestone
and
then
go
from
there.
I
can
and
then
I
can
try
to
regurgitate
my
my
thinking
on
infrastructure,
but
James.
You
can
take
it
away.
First,.
F
Yeah,
we'll
be
fine,
so
16
5,
but
I
guess
you
had
are
a
lot
of
issues.
F
Our
theme
for
this
Milestone
is
what
customers
on
board
themselves,
that
aligned
with
our
OTR,
releasing
product
analytics
and
then
open
Beta,
that's
one
of
the
product
objectives
for
the
quarter,
and
it's
definitely
a
stretch
on
the
engineering
side
objective
for
the
quarter,
some
of
the
the
to
Do's,
I,
guess
or
specific
tasks
there,
making
product
analytics
that
opt-in
of
manual
onboarding
item
Max
has
already
been
starting
to
work
on
that
from
front-end
perspective,
resolving
any
of
those
PT
usability
or
functionality
bugs
I
think
that
most
of
that
lists
have
already
burned
through
as
well,
are
being
assigned,
and
then
the
big
one
stress,
testing
the
cluster
and
that's
I
think
we're
well
on
our
way.
F
There
kudos
to
Dennis
we're
getting
like
Health
Cloud
set
up
and
nothing
is
on
fire.
Yet,
as
far
as
I
know,
I
haven't
heard
about
data
center.
You
know
problems
and
then
the
last
one
here
is
a
stretch
goal
around
implementing
the
event
rate
limits.
You
need
to
think
about
as
not
just
we
on
board,
but
as
we
start
to
ramp
this
up
and
make
it
an
opt-in
process
for
the
ultimate
South
customers.
F
How
do
we
put
some
limits
in
place
so
that
we
don't
impact
the
folks
already
on
board
but
prevent
folks
from
abusing
whether
intentional
or
not
the
analytics
stack?
So
there's
an
issue
to
define
the
usage
limit.
I've
put
some
numbers
in
there
to
start
the
conversation,
but
those
are
going
to
be
the
big
things
to
focus
on
and
deliver
within
the
Milestone.
Okay,
we'll
also
start
down
that
usage
quarter
path.
Discussion.
We
already
had
we're
well
on
our
way
there
and
that'll
be
something
we
want
to
introduce
during
the
beta.
F
So
that's
where
it's
going
to
start
to
get
offensive.
How
many
events
are
they
sending?
Are
they
getting
value
out
of
those
and
I
think
that
the
one
that's
not
built
in
here,
but
is
a
pullover
or
a
carryover
from
16-4,
is
wrapping
up
the
great
work?
That's
our
happened
on
the
dashboard
designer
I
think
there's
a
couple
of
little
things
there,
but
it's
close
Dennis.
What
did
I
miss
from
our
conversation
earlier
today,
I'm
going
to
try
to
multicap
too.
A
Much
yeah
I,
don't
think
you've
missed
anything
that
was
just
thinking
for
usage
limits.
A
Oh
that's
a
tough
one,
especially
with
the
events
per
second,
because
I
guess
are
we
defining
the
bounds
of
usage
limits
for
the
beta
to
start
or
just
in
general
and
like
I,.
F
Guess
just
in
general
and
we
don't
necessarily
need
to
publish
them
yet
you
can
get
to
that
point,
but
this
is
again
more
discussion
of
what
do
you
already
see
today?
What
do
we
think
this
back
can
handle?
Realistically?
What
should
we
expect
from
a
good
actor
spending
a
number
of
events,
and
do
we
apply
this
at
a
collector
level
or
a
customer
level
or
an
app
ID
level,
rather.
A
Yeah
that
that
gets
particularly
interesting
because
it
also
places
an
emphasis
potentially
on
usage
quota
and
I
mean
it
was
a
proposal
and
it
was
an
idea
but
I
apologize
for
calling
out
a
potential
number
for
the
initiative
that
has
is
confidential
yeah.
The
the
the
important
thing
is
like
we
have
to
be
able
to
put
the
controls
in
place
and
for
us
to
do
that.
A
A
But
if
anything
to
your
point,
it's
about
defining
it
so
I
would
wager
we
have
to
do
a
little
bit
more
information
gathering
before
we
have
a
better
idea
about
that,
because
I
guess
just
initial
Impressions
is
I.
Don't
it
doesn't
make
sense,
I
mean
that's
quite
substantial.
That
means
you're
you're.
E
A
Hundreds
of
millions
of
events,
if
not
billions,
of
events
per
day,
if
it's
per
second,
so
that's
definitely
sensible.
I
guess
yeah
I
just
have
to
think
on
that
more
I
guess
to
related.
The
stack
has
been
doing
quite
well
to
the
point
where
I
want
to
enable
enable
the
entire
gitlab.com
instrumentation,
because
the
numbers
just
aren't
adding
up
right
now
and
I'm
trying
to
understand
why
and
the
stack
hasn't
fallen
over.
F
Cool
so
you're
making
great
progress,
Dennis
and
I
were
talking
earlier
today
about
how
far
ahead.
We
are
that
we're
starting
to
focus
on
polish
things,
and
maybe
we
should
leave
a
little
rough
around
the
edges
to
get
customer
feedback
on
so
I
mean
just
great
work,
everyone
of
seeing
things
picking
them
up.
Thinking
about
the
customer
experience
and
trying
to
make
it
better
even
before
we
have
those
customers.
So
don't
stop
what
you're
doing
we'll
just
try
to
provide
better
direction
to
you
of
where
the
where
the
focus
should
be.
A
Definitely
and
I
think
I
have
a
good
helping
of
gender
topic
to
kind
of
speak
to
both,
but
I
had
something
about
abbreviating
numbers
on
the
single
stab,
but
we
can
go
ahead
and
move
on
that
because
we're
recording
a
little
bit
out
of
time.
The
main
thing
I
wanted
to
call
out
and
James
was
talking
about
it,
so
we
were
having
a
discussion
and
I
think
it
was
it's.
If
it's
not,
let's
make
it
as
obvious
as
we
can
now.
A
Given
the
you
know,
the
delays
we've
had
in
getting
out
to
experiment
phase
and
from
an
infrastructure
and
logistical
point
of
view,
because
we've,
obviously
you
know,
fulfilled
what
we
wanted
to
build
from
a
feature
development
perspective
months
ago.
Given
that
we're
now
really
polishing
and
adding
things
like
error,
validation,
saving
changes,
properly
and
being
able
to
retry
queries,
and
things
like
that.
A
So
we're
really
definitely
in
that
that
polish
phase
I
I'm
trying
to
be
as
transparent
about
what's
floating
around
in
terms
of
the
infrastructure
side
of
things
and
I
know,
a
couple
of
people
have
reached
out
in
terms
of
how
they
can
help
with
that.
So
I
initially
started
off
this
long
agenda
list
about
all
the
different
topics
that
are
flying
around
with
infrastructure
as
potentially
candidates
that
we
could
swarm
on,
but
some
of
them
aren't
ready
yet,
and
some
of
them
really
need
some
discussions.
A
I
need
with
infrastructure
to
be
like
I
know
what
the
solution
looks
like,
but
what
can
we
do
with
the
help
that
we
have,
and
we
can
speak
more
to
that
there,
but
we
don't
have
a
dedicated
SRE
counterpart
right
now
and
so
I'm
trying
to
figure
out
what
our
options
are
there.
There
are
certain
issues
that
I
think
we
could
potentially
run
with
so
like
getting
someone
just
getting
project
the
people
running
Project,
Runway,
aware
of
what
we're
doing
and
having
them
engage
with
us
is
one
getting
the
cube,
store.
A
Architecture
working
properly
is
another
and
then
we're
already
looking
at
usage
quota
as
well,
but
take
a
look
at
that
list
of
stuff
and
if
you
think
you
can
help
or
help
run
with
the
or
help
better
Define
that
please
feel
free
to
reach
out.
Basically,
we've
we've
unofficially
launched
an
experiment
phase.
A
There's
a
infrastructure
process
to
get
us
approved
as
far
as
like
being
officially
an
experiment
phase,
but
I
really
want
us
to
make
sure
that
we
can
catch
up
from
the
infrastructure
and
kind
of
logistical
side
of
things
so
that
we
know
we're
ready
for
beta
and
I.
Think.
A
lot
of
that
is
what
James
has
already
called
out
in
in
the
themes
for
this
milestone.
A
Quite
promising,
to
be
honest,
I
think
it's
it's
going
to
be
more
of
the
instead
of
asking
is
the
stack
gonna
fall
over
with
something
like.com
is
it
might
be
more
of
a
question
of
like
how
can
we
make
it?
How
can
we
ensure
it's?
It's
got
enough
Headroom,
but
is
also
running
efficiently
as
possible,
so
that
we're
not
over
committing
resources
and
causing
it
to
scale
inefficiently
in
other
ways.
A
But
anyways
have
a
read
through
that.
If
that's
something,
if
there's
anything
there,
that
you
think
you
can
help
with
please
reach
out
will
certainly
be
easier
than
than
you
know.
It'll
be
a
lot
better
if
we
can
have
more
people
engaged
there,
but
to
James
point.
If
there's
any
question
on
what
to
work
on,
please
reach
out
cool,
there's
a
lot
of
rambling,
a
couple
of
housekeeping
items.
I
just
wanted
to
see.
A
If
anyone
had
any
questions
on
what's
changed,
there
had
been
a
lot
of
developments
with
the
analytics
stack
and
the
configurator
to
we
did
things
like
dropping
coffee
at
a
table
support
we
added
a
new
thing
called
Vector,
which
we
touched
on
and
we
support
clickhouse
Cloud,
so
I
know
we've
been
like
you
know:
yay
we're
there,
but
I
just
wanted
to
see.
If
anyone
had
questions
about
like
what's
what's
changed
or
anything
like
that
I
know,
we've
also
had
a
discussion
around
making
Vector
the
standard,
but
yeah.
A
Cool
other
than
that,
the
on
the
other
topic-
I,
don't
think
I've
touched
on-
is
that
we
will
be
getting
a
backfill
for
for
Allen.
So
if
you
have
any
referrals
for
back-end,
Engineers
I'm
happy
to
receive
them,
I'm
working
with
our
recruiter
to
open
up
the
backfill
I
will
say
I'm
looking
for
backing
engineer
with
some
infrastructure
experience,
so
they
we
can
lean
on
their
experience
to
kind
of
help.
Make
more
progress
on
some
of
these
topics.
But
anybody
can
engineer
is
also
fine.
A
A
Awesome
I
was
going
to
see
everybody
loves
the
discussion
today
and
have
a
good
rest
of
your
week.
Don't
forget
Friday
and
his
family
and
friends
day
so
have.