►
From YouTube: Applied ML weekly team meeting Sep 8 2021
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
A
Yeah,
so
I
I
put
the
summary
in
the
meeting
notes
for
the
meeting.
I'm
actually
it's
worth
reading
through
these,
so
milestone,
one
is
done
yay.
In
addition
to
the
architectural
plan
being
vetted
and
published
in
the
handbook,
we
migrated
unreview
from
azure
to
gcp.
We
run
it.
We've
run
it
on
a
gitlab
project.
Try
giddily
in
particular
and
receive
feedback
on
it
from
the
development
team.
We've
started
work
on
milestone.
Two
other
progress
is
meehau,
moved
back
to
the
threat.
A
Insights
team
was
a
better
match
for
his
interest,
which
is
great
for
me.
How
and
great
for
us
we
are
working
on
a
backfill
wreck,
not
having
a
second
engineer
on
the
team
with
alexander
and
one.
That
already
knows,
the
gitlab
good
code
base
will
make
it
challenging
but
of
course,
not
impossible
to
meet
our
goals
around
time
frames
for
milestone
two,
but
we're
going
to
do
our
best.
A
While
we
continue
to
work
to
hire
and
get
that
rec
listed
both
internally
and
externally,
and
as
we
work
to
dog
food
on
review
at
gitlab,
we
were
always
predicting
that
we'd
hit
up
against
the
channels
that
reviewer
roulette
is
not
part
of
the
gitlab
product
and
uses
data
and
workflows
that
are
unique
to
gitlab
and
not
used
and
in
some
cases,
not
even
desirable
for
customers.
Things
like
custom,
stat,
statuses
of
users
and
time
zone
of
the
user
information
from
team.yaml,
which
is
not
something
that's
not
part
of
the
product.
A
That's
part
of
the
get
lab
engineering
workflow,
so
we
will
address
in
the
short
term
via
making
unreview
recommend
reviewers,
but
not
assign
them,
so
it
doesn't
affect
and
conflict
with
current
reviewer
roulette
in
the
future.
We
may
actually
want
to
integrate
integrate
portions
of
the
viewer
roulette
like
features
or
roulette
like
features
not
review
or
let
itself
into
the
product
itself,
so
that
our
customers
can
benefit
from
the
combined
features.
So
that's
my
my
quick
summary.
What
are
you?
What
are
your
thoughts
on
that
alexander.
B
B
A
Yeah
good
good,
good,
good
thoughts.
We
did
get
a
lot
of
good
feedback,
but
they
seemed
to
be
worried
and
surprised
like
worried
that
we
were
gonna
go
faster
than
they
were
ready
for
us
to,
and
we
weren't
planning
to
do
that.
But
you
know
so
be
it,
but
also
they
were
surprised.
I
think
we
could
have
helped.
People
are
always
going
to
be
surprised,
we're
always
so
busy
as
a
as
an
organization,
but
we
could
have
done
better
on
the
surprise
side.
So
it's
point
well
taken.
We
should.
B
But
the
best
thing
that
we
don't
affect
the
current
workflow,
because
we
just
add
and
review
as
an
experimental
feature,
and
we
just
put
the
recommended
team
members
the
into
the
comments
without
assigning
them
directly
to
the
merch
request.
So
that's
why
it's
up
to
the
author
of
the
merch
request
or,
for
example,
maintainers
who
will
review
emergency
quest?
We
only
we
only
want.
We
just
want
to
make
this
recommendation
and
then
we
will
collect
the
will
collect
the
assignments
automatically
just
to
collect
feedback
and
estimate
metrics
and
to
improve
the
model.
A
Yeah
and
just
saying,
hey,
we've
tried
it
with
giddily
we've
gotten
some
feedback
already
we'd
love
more.
So,
if
you
can
add,
add
that
to
engineering
we
can
review
that'll
get
the
information
out
more
to
more
engineers.
A
And
last
item
we
have
on
the
agenda,
is
you
know,
alexander?
You
mentioned
you're
taking
time
off
next
week
we
have
a
team
of
two
with
you
out
next
week
and
mon
out
till
the
end
of
september.
You
know
so
we're
going
to
be
on
pause
primarily
next
week,
I'm
going
to
keep
things
moving
on
the
recruiting
process
in
mon's
absence,
but
we'll
start
planning
out
milestone,
2
in
more
detail
when
you
return
week
after
next.
A
Yeah,
it
sounds
good
we
want
to
be
we
need
to.
I
need
to
look
it
over
again.
Myself
determine
here
were
the
goals
here,
the
things
that
we've
learned,
that
we
were
expecting
here
things
we
learned,
we
weren't
expecting
and
then
look
at
all
the
different
things
that
we
didn't
do
also
in
milestone,
one
that
we
pushed
into
milestone
two
and
get
lots
of
feedback
and
direction
from
taylor,
of
course,
and
then
the
kai
and
the
others
you're
gonna
be
working
with
to
figure
out.
A
What's
next
so
taylor,
you
made
you
wanna
verbalize
number
three.
C
A
I
think
we're
good.
I
did
an
update
in
the
issue.
I
nudged
everyone
to
read
it
in
advance,
we'll
see
who
actually
does
some
will
have
the
time
to
do
that,
so
it
won't
but
yeah.
I
think
we're
in
good
shape.
C
Okay,
I
figured
tomorrow,
we
also
should
show
the
demo
of
unreview.
I
think
that
would
be
insightful
to
folks
and
we
might
get
some
good
feedback.
A
Yeah,
I
think
you
show
the
example
yeah
of
the
recommendation
from
italy
like
you're
mentioning,
and
I
think
also
even
though
we
may
not
use
it
from
definitely
won't
use
it
as
is,
but
we
may
we
may
not
use
it
really
as
the
the
ui,
the
under
review
ui,
I
think
that's
that's
neat
to
see
as
well.
A
Right
cool
that
so
tomorrow,
in
that
meeting
I'll
I'll
go
over
the
summary
and
then
and
then
we'll
be
great
to
do
the
demo
in
fact
actually
alexander
my
well
yeah,
I
think
then
we
can
just
do
a
demo
live.
C
Awesome,
the
only
other
thing
I've
got
is
that
I
am
officially
throwing
my
hat
into
the
ring
for
the
model.
Ops
pm
roll,
so
I
applied
last
night.
I
have
gotten
good
feedback
from
folks
already,
so
we
all
may
have
a
pm
sooner
than
later,.
A
C
If
you
see
mon
online
like
tell
her
to
enjoy
her
time
off,
she's
already
come
online
a
few
times.
A
That's
a
good
point
and
we'll
we'll
keep
an
eye
on
that
next
week
with
alexander
as
well
as
alexander.
We
don't
expect
you
to
see
you
online
this
week.
Take
your
time
off
and
get
recharged
enjoy
your
title.
Absolutely.