►
From YouTube: 2019 06 27 Memory Team Meeting
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).
B
B
Like
everybody
see
the
screen,
alright,
okay,
I,
don't
think
andrew
has
joined
us,
but
we
had
a
meeting
a
couple
weeks
ago
or
we
talked
about
puma
infrastructure
memory.
Team
topics
I've
run
through
this
doc
a
couple
times,
I
didn't
see
anything
that
was
not
covered
with
issues
are
assigned
to
folks,
so
does
anybody
know?
Are
there
any
outstanding
issues
and
even
Andrew
said
he
has
I
had
a
chance
to
look
at
this
or
there's
he
didn't
mention
anything
pressing
right
now.
C
Don't
think
that
we
have
a
mic
matter:
blockers
I,
think
that
they're,
like
I,
noticed
some
inconsistent
matrix,
which
is
like
not
related
to
permits,
but
also,
but
pretty
much
throughout
our
Prometheus
metrics
monitoring.
That
would
really
help
to
get
into
stripe.
I
think,
even
though,
like
we
have
these
items
open,
there
is
like
no
clear
like
deadline
and
my
action
point
on
who
is
doing
what
and
executing
parts,
and
this
is
probably
like
the
most
missing
piece
at
this
moment.
B
B
There's
not
like
one
overwhelming
theme
on
this
work,
it's
kind
of
scattered
everywhere,
and
a
lot
of
these
will
be
addressed
by
it
once
we
get
done
with
the
Puma
work
right,
so
is
it
something
that
we
should
address
after
we
get
Puma
shipped
in
twelve
to
we
hit
that
goal
or
enough
of
these
related
that
we
need
to
start.
We
need
to
redress
with
Andrew
and
spend
some
time
scheduling
this
workout
I
would.
C
I
would
talk
funny
later
than
half
of
these
items
here.
It's
like.
We
need
infrastructure
engineers
to
start
implementing
these
things
and
and
I
bring
that
the
second
half
is
Vera
asked
focusing
on
best
ink
and
I
figure
encountered
like
some
performance
benchmark.
What
John
did
start
the
issue
some
time
ago
to
actually
have
some
understanding
of
the
baseline
and
when
Puma
does
performer
and
English,
even
since
it
does
not
perform,
because
it
seems
that
right,
it's
good
enough,
but
we
don't
have
like
conclusive
data,
exactly
what
we
were.
C
What
we
gonna
like
how
system,
how
we
expect
the
system
to
be
safe?
Why
we
enabled
that,
let's
say
for
everyone
on
github.com-
and
this
seems
like
that,
like
the
two
biggest
missing
pieces
like
I
stink,
like
performance
testing
scenario,
testing
like
load
testing,
as
it's
mentioned
here
and
the
second
one
is
really
like.
We
need
people
that
gonna
start
making.
C
These
changes
on
the
staging
Connery
is
gonna
like
like
an
able
that
they're
gonna,
like
figure
out
what,
as
what
other
metrics
they
need
and
what
they
don't
understand
about
the
system
and
we
help
them
fill
the
gaps.
We
we
are
right
now
working
like
with
the
quality
team
on
the
testing
part,
but
there
is
nothing
happening
on
the
infrastructure
side.
This
is
probably
like
the
biggest
thing
that
is
right
now,
BC.
B
Okay,
maybe
so
here's
what
I'm
thinking
I
can
ping
Andrew
to
have
them
join
us
next
week,
Tania
and
grant
I
don't
know
that
they
were
well
grant
for
sure
wasn't
in
on
previous
discussions
about
this
document.
C
Do
you
have
some
recommendations
on
who
should
we
reach
from
an
IT
infrastructure
team,
because
it
seems
like
the
Pullman
Strike,
the
top
priority
for
like
and
I've
met
an
infrastructure,
team
and
I?
Think
that,
like
it
would
be
really
helpful
if
we
could
have
someone
from
the
other
side
be
like
very
deeply
involved
in
our
meetings
about
the
Puma
there.
B
C
Andrew,
it's
like
a
distinguished
engineer
from
the
infrastructure
team,
but
recently,
and
you
were
like,
what's
really
overwhelmed
if
I,
if
I
think
grants
also
I
would
really
look
at
the
agile
like,
let's
say
approval,
but
like
we
need
someone
that
would
really
focus
hands-on
on
these
items.
Okay,.
D
C
E
On,
as
for
me,
I
started
to
make
make
myself
familiar
with
issues
you
shared
the
two
ones
also
create
commented
on
Puma,
but
as
we
discussed
with
Camille,
we
are
probably
not
going
to
handle
it
right
from
the
start.
So
I
guess
I'm.
Looking
into
these
two
issues,
one
with
CI
another
with
omnibus
so
I'm
going
to
comment
on
any
of
that
which
one
I'm
going
to
start.
First.
F
For
my
part
or
fix
the
the
issue,
six,
two,
two:
seven
zero.
If
we
don't
need
to
handle
profile,
URL
well,
then
I
think
MRI
is
ready
for
review,
but
I
also
prefer.
We
have
a
pattern
for
other
issues,
because
I
feel
like
when
I
work
on
some
issue
and
some
time,
I'm
blocked
because
I'm
lacking
of
some
information.
In
my
day
time,
then,
if
I
have
several
issues
in
the
pipeline,
I
can
go
to
look
at
other
issues,
so
that
will
be
make
I.
F
B
Yeah
something
camille
mentioned
in
the
chat,
we'll
find
some
more
issues
to
pre-assigned
to
folks
and
when
we
plan
for
12
we'll
make
sure
that
there's
plenty
assigned
for
folks
for
the
whole
milestone.
So
then
think
Camille's
point
that
can
be
changed
whenever
they
need
to
just
like
for
an
idea
was
changed
from
Camille
de
to
Yan
today,
and
the
reason
I
mentioned
that
one
in
the
ticket
for
Camille
I'm,
sorry,
14,
you
and
Alexei
is
490-
is
pretty
straightforward.
It's
a
setting
that
needs
to
be
updated
and
it's
pretty
well
defined
in
the
ticket.
B
C
So
I
would
say,
like
I,
just
people
just
figure
out
between
you,
like
who
just
gonna
figure
out
like
what
work
next
I
think
it
will
be
like
the
most
efficient
way
for
you
to
like
to
talk
and
comment
on
the
issue
about,
like
the
ideas
and
proposals
on
like
how
to
approach
study.
It
fan
of
I
think
forcing
to
affinity
of
like
you
what
you
are
working
so
far
on.
B
See
recruiting
update
so
I've
asked
for
five
candidates
to
be
screened.
Actually,
I've
asked
for
six,
but
one
of
them
was
rejected
pretty
quickly.
Two
of
them
are
being
screened.
Right
now
are
scheduled
for
screening
and
one
of
them
pass
through
to
a
technical
interview.
So
that's
just
for
engineering
for
a
product
manager
for
the
memory
team.
We
have
a
really
strong
candidate,
Chen
and
I
worked
with
in
the
past
I
interviewed
earlier
this
week.
B
She
interviewed
with
product
management
has
an
exercise
that
she
means
to
return
I
think,
probably
tomorrow
they
have
a
to
have
an
interview
and
then,
four
days
later,
the
candidate
has
to
return
their
vision.
It's
part
of
the
exercise
so
get
more
feedback
later
on
this
week.
Hopefully
she
moved
forward.
Larissa
is
fantastic,
though
any
questions
on
recruiting.
B
B
A
Yeah,
so
just
a
general
hazop
that
the
type
of
organization
is
shooting
to
move
to
the
seedy
model
right
now
the
dirac
aim
is
still
stressed
strengthening
what
was
missing
the
the
development
process.
There
are
still
many
errors
or
failures
there,
but
once
that's
sorted
out,
we
are
going
to
move
to
the
city
model,
so
that
may
require
a
lot
of
practice.
Changes
in
our
process.
A
So
at
this
point,
I
haven't
seen
any
articulated
the
process
how
to
move
to
there,
but
we
can
think
about
it
and
geo
team
is
going
to
practice
that
first,
so
they
are,
they
have
something
laid
out
already
and
I
asked
the
Geo
team
to
document
their
learnings
in
the
lessons.
So
we
can,
you
know
we
don't
need
to
repeat
the
same
mistakes.
B
All
right,
we
had
the
12-point
Oh
retro
earlier
this
week
and
memory
team
didn't
exist
for
12
by
now,
so
we
didn't
have
to
deliver
anything
I
would
encourage
you
to
take
a
look
at
the
previous
teams
retro.
So
we
know
what
what
we
have
to
deliver
for
twelve
one
I've
already
set
up
the
async
retro
space
for
our
team,
so
that
will
automatically
get
tickets
created.
We
do
have
one
set
up
already
for
twelve
one.
We
can
start
contributing
to
it.
B
I
would
recommend,
as
things
come
up,
because
a
month
is
a
long
time
like
things
that
go
well
things
that
don't
go
well
just
start,
adding
them
to
the
ticket
as
you
encounter
them,
rather
than
waiting
till
the
end
of
the
month.
Because
often
we
forget
something
awesome
that
happened
and
it's
hard
to
forget
things
that
are
terrible.
That
happened,
but
please
update
the
retro
issues
frequently
and
feel
free
to
add
whatever
you
want,
and
then
those
will
all
get
rolled
up
into
the
company-wide
retro
at
the
end
of
the
month.
B
As
for
the
twelve
to
schedule,
it's
not
ready,
there's
nothing
to
preview
yet
so
Camille
and
I
are
gonna
work
on
that
tomorrow
and
try
and
get
more
specific
tasks
for
this
team
and
especially
get
a
timeline
for
when
we
get
Puma
out
to
production.
The
goal
is
to
have
it
live
in
production
by
the
end
of
twelve
I.
B
Don't
know
how
feasible
that
is
yet
because
there's
going
to
be
a
lot
of
dependencies
and
potential
blockers,
but
we're
in
trying
lay
out
a
timeline
to
get
to
my
live
in
production,
on
12
to
any
questions,
all
right
and
then
the
last
thing
on
labels.
So
it's
one
of
my
tasks
to
look
at
the
labels
of
issues
on
a
regular
basis
to
make
sure
that
we
have
the
right
labels
on
what
we're
working
on
on
mrs
make
sure
they
are
identified
correctly
and
memory
teams
doing
great
no
concerns
there.