►
From YouTube: 2021-05-13 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).
A
B
E
E
I
think
this
is
everyone
lays
out
and
oh
it
takes
out
all
right
who
wants
to
open
up
the
kanban.
F
Yeah,
so
I
did
a
few
more
sort
of
maintenance
tests
yesterday
did
a
follow
up
on
the
bug
fix
for
the
grpc
context,
bridge
bug
that
was
reported
a
week
ago,
or
so
it
was
just
a
log
message
that
wasn't
needed
anymore,
and
then
I
made
a
bit
more
progress
on
a
dock
that
I'm
writing
to
describe
some
of
the
things
related
to
our
internal
integration
strategy
unless
we
have
a
meeting
with
alex
in
15
minutes.
But
I
feel
as
if
thought
of
been
a
better
meeting
to
wait
for
my
doc,
but
we'll
see.
E
Yeah,
it's
it's
just
an
informal
touch
base
because
I
haven't
had
a
chance
to
chat
with
you
and
alex
has
some
changes
on
his
send
in
regards
to
baggage,
and
I
just
figured
that
it
would
be
a
good
time
for
us
to
set
up
a
recurring
meeting
so.
F
E
Today,
okay,
what
else
do
we
got
on
here
for
ya?
There's
various
orr
items,
but
those
are
kind
of
low
low
priority,
more
simple
apps
and
playgrounds
low
priority
yeah,
not
a
huge
amount,
any
collets
from
anyone
else.
F
D
D
Yeah,
so
for
my
end,
I
did
a
little
bit
of
work
on
the
javascript
api
doc
migration
this
morning,
but
it's
probably
gonna
be
a
little
bit
of
a
task,
so
just
tabled
that,
since
it's
not
really
that
high
priority,
I
I
don't
think
that
the
security
vulnerabilities
act
like
it's
just
not
a
debt
dependency,
so
I
couldn't
impact
any
customers,
and
then
I
was
working
on
the
release
for
the
node
sdk.
D
So
just
waiting
on
sandra
your
approval
and
then
I
can
probably
get
that
out
and
I'll
also
be
releasing
a
patch
of
the
job
sdk.
Since
we
had
a
threading
issue
in
the
latest
release
that
we
just
did
a
second
customer
report,
a
possible
issue
with.
So
I
want
to
get
that
out
soon
and
then
I
also
just
been
spending
a
little
bit
of
time,
updating
our
sample
app
and
the
aws
hotel,
js
repo
to
the
latest
versions
of
the
upstream
sdk
and
api.
D
E
G
Yeah:
okay,
for
me,
semi
about
the
ota
domain
instrumentation
for
london,
which
will
be
donald
trump
and
I
think
prashan
is
taking
a
look
at
it.
And
yes,
only
the
terraform
configuration
files
and
sample
apps
to
the
upstream
and
the
downstream
before
the
number
and
then
the
repo
and
the
synthetic
the
csd
test
compass
has
passed
and
also
responded
on
the
security
findings
planning
tickets
and
we
are
waiting
for
a
production
about
the
united
states
interest
and
also
was
almost
working
on
customer
tickets.
D
Are
all
these
still
under
review?
G
D
There
yeah
and
also
the
is
the
abstract
review
done.
Oh
yeah
right,
you
can
check
that's
done.
G
E
E
Yeah
they
will
literally
like
send
to
us
and
make
us
pull
down
our
release
marketing
materials.
If,
if
this
isn't
closed
out
before
we
launch
they,
they
do
crazy,
crazy
things.
If
security
reviews
aren't
properly
closed
out,
so
just
want
to
make
sure
that
your
security
review
is
officially
closed
it
out
only
when,
on
the
left
hand,
side,
all
the
green
check
marks
are
done.
E
If
the
green
check
marks
aren't
done
yet,
then
then
we
still
have
work
to
do,
which
usually
involves
us
chasing
down
apsec,
which
is
why
I
was
hoping
that
we
could
get
it
closed
out
this
week,
because
elizabeth
is
already
handing
off
to
the
new
engineer-
and
this
is
due
at
the
end
of
the
month.
So
if
we
could
follow
up
with
elizabeth
to
get
it
closed
out
great,
otherwise
we
have
to.
G
Yeah
yeah
yeah,
as
far
as
that,
if
we
can
continue
working
on
this
because
it's
right
from
the
university
this
project-
and
it
might
be
like
easier
for
her
to
onboard
that
issue.
Looking
into
the
new
engineer
so.
H
She's
transitioning
on
may
14th.
E
Oh
geez,
okay,
yeah,
can
you
sit
below?
Can
you
send
out
an
email
to
nazar
myself
and
elita
and
just
say
hi
to
the
new
security
engineer
and
also
tell
him
that
we
have
a
release
at
the
end
of
the
month
that
elizabeth
didn't
get
a
chance
to
close
out
and
if
we
can
get
that
closed
out?
H
Okay,
so
for
me
I
finished
up
the
investigation
on
the
aws
tracing
control
plane
package
for
switching
out
the
allow
listing
class
to
using
service
principle
had
a
question
about
performance
impact
of
doing
some
string
operations
like
splitting
and
matching
so
talk
to
bilal
about
it,
and
he
he
showed
me
actually
the
aws
profile
how
to
use
aws
profiler,
and
he
said
I
could
then
doing
implementation.
We
could
just
merge
it
to
alpha
and
then
it
address
provider
would
pick
it
up
and
see
the
performance.
H
So
I
documented
all
of
this
on
that
ticket
and
resolved
it
will
open
a
new
one
per
implementation.
Part
and
right
now,
I'm
investigating
the
same
for
the
data
plane
package
other
than
that
took
a
a
quick
overview
on
borus
pr
for
lambda
support
I'll
be
taking
another
pass
at
it
later
today.
D
Just
a
quick
question:
should
I
go
ahead
and
do
the
node
release
after
another
meeting
after
this
still
five,
but
should
I
get
out
of
the
way
quickly
at
five
or
should
we
wait
until
tomorrow?
I
I'm
okay
with
doing
it
at
five,
but
just
wanna
get
the
team.
E
D
E
A
E
Hey
so
I
marked
down
dennis
is
optional.
He
might
come
every
so
often
to
kind
of
see
where
we're
at
or
whatever
but
yeah.
So
I
just
figured
we
could
book
a
meeting
this
wednesday
and
just
kind
of
sync
together
on
where
we're
at.
What's
going
on
what
the
next
steps
are
and
any
call-outs
concerns
and
shifting
sands
and
all
that
kind
of
stuff
alex,
I
know
you're
busy
on
your
side
dealing
with
the
entire
baggage
thing
nazar
doesn't
want
us
to
spend
time
on
baggage.
E
So
if
that
gets
somehow
muddled
into
the
w3c
work,
then
that's
going
to
be
an
issue.
Oh
yeah.
I
forgot
I'm
so
sorry,
william.
I
forgot
that
I
invited
you.
E
Yeah,
so
basically,
I'm
hoping
that
you
can
help
contribute
to
the
wc
3c
and
integrations
project
as
you've
worked
on
integrations
before
and
and
yeah.
Hopefully,
you
and
anarag
can
do
a
little
bit
of
brainstorming
on
rug.
You
can
definitely
use
william
to
help
out
on
input
or
docs
or
or
whatever,
but
I
figured
that
we
could
have
kind
of
like
a
main
ownership
free
for
for
honorable
and
then
well.
E
You
can
just
kind
of
be
involved,
see
if
there's
any
like
follow-up
items
or
reach
outs
that
we
need
to
do
or
things
like
that
with
other
teams.
I
know
honorable
be
kind
of
difficult
for
you,
considering
the
time
zone,
differences
to
get
information
if
you
need
it,
so
I'm
hoping
that
we
can
all
kind
of
work
together
and
collaborate
on
this
effort
together
and
what
we
can
do
with
integrations
and
w3c.
This
course.
E
Perfect
so
honorable,
you
wanna
tell
us
what
you've
been
kind
of
working
on
lately
and
then
we
can
just
pretty
much
have
like
a
mini
stand
up.
F
Okay,
so
last
I
chatted
with
alex,
I
think,
last
friday
or
so
last
friday
in
my
time,
and
he
suggested
I
should
write
a
doc
sort
of
outlining
where
we
want
to
get
to
and
some
of
the
steps
to
get
there,
and
so
I've
been
working
on
the
doc
this
week
sort
of
lazily.
But
I'm
going
to
try
to
finish
that
up
today.
F
Just
together
like
there
will
be
multiple
stages
as
well,
and
it's
not
like
every
stage
is
going
to
get
us
to
the
final
goal,
but
I
just
want
to
make
sure
that's
written
down,
so
it's
easier
for
people
to
process
what
we're
doing
and
that's
about
it.
I
mean
I
still
like
I
feel
as
if
code
wise,
like
it's
mostly
like
the
basic
components
of
how
to
write
a
propagator
sort
of
already
done,
and
it's
more
just.
We
have
to
figure
out
how
to
roll
this
out
to
people.
F
I
think
it's
mostly
identifying
I
mean,
as
you've
said,
like
every
team
does
seem
to
have
some
unique
requirements.
I'm
hoping
they're
not
that
unique,
because
I
do
want
to
find
some
common
grounds.
We
don't
have
to
implement
every
single
time,
but
it's
identifying
what
those
are
and
seeing
if
it
is
actually
possible
to
at
least
share
some
code,
if
not
have
a
proper
framework
that
everyone
can
use.
E
Yeah
yeah
absolutely
so
we
have
the
huge
integration
stock
wiki.
I
don't
know
if
you've
been
picking
through
that,
but
that
has
a
lot
of
the
of
the
links
to
either
or
the
code
that
we
own
or
the
code
that
other
teams
own.
F
E
Awesome
yeah,
so
I'm
wondering
I'm
wondering
if
we're
planning
on
vending
something
or
we're
planning
on
having
you
know,
better,
better
documentation,
easier
integrations
and
how
this
kind
of
fits
in
between
w3c
and
integrations
mozar
is
kind
of
treating
them
as
two
separate
things,
as
is
dennis,
and
I
want
to
kind
of
have
that
conversation
with
you
on
how
how
do
you
see
this
is?
Is
ww3c
separate
from
integrations?
E
F
Mean
so
they're
the
same
because
context,
propagation
is
context
propagation,
whether
it's
x-ray
or
w3c,
or
whatever.
It's
still
the
same
code.
For
the
most
part,
you
just
have
to
configure
what
basically
headers
or
what
state
you're
looking
at
when
you
propagate
and
that's
actually
relatively
easy,
if
you're
actually
being
able
to
move
the
context
around.
So
I
like
w3c
baggage
or
even
x-ray,
like
all
these
things
fall
under
the
same
bucket
of
work
and
yeah.
I
definitely
hope
to
vent
some
code
that
can
do
that
for
everyone
at
least
almost
everyone.
F
E
E
Yeah,
I
guess
I'm
just
kind
of
looking
for
a
little
bit
more
concrete
aspects,
because
alex
is
saying
that
if
we
need
to
own
some
sort
of
baggage
framework,
that'll
have
to
fit
a
couple
of
use
cases
with
s3.
I
believe-
and
you
have
a
short
list
of
services
right.
A
A
F
You
know
I
mean,
I
think
the
point
there
is
like
whether
it's
baggage
like
it
could
even
be
a
different
header,
really
like
I'm
definitely
planning
on
using
the
hotel
code
base,
because
most
stuff
is
implemented
already,
including
w3c,
propagation
or
whatnot,
but
like
it
has
an
object,
called
context
and
what's
inside
there
is
just
what
you
choose
to
put
inside
there,
it
could
be
w3c
or
baggage,
even
lineage
or
whatever
how
you
use
that
it
could
be
sort
of
up
to
the
services,
but
the
hard
part
is
making
sure
that
context.
F
F
A
I
think
it's
a
good
definitely
to
keep
that
in
perspective,
that
you
know
that's
pretty
universal
you're
explaining
one
thing
might
help.
I
don't
know,
and
maybe
under
you
already
familiarize
yourself
with
other
integrations
like
it
may
be
helpful
to,
for
example,
just
look
at
how
sd
integration
is
done.
A
A
I
know
I've
heard
again.
I
did
not
confirm
this,
but
I've
heard
that
they
come
even
compressed,
like
x-ray
header,
just
to
try
to
save
on
some
lights,
so
that
may
a
little
bit
provide
some
input
for
this
universal
context
operations.
I
mean,
I
don't
know
at
some
point.
They
will
need
to
syria,
like
this
allies,
but
right.
F
The
thing
to
look
through
that
is
to
confirm
what
the
current
state
is
still
thinking
like.
If
the
mechanism
can
be
standard,
even
if
the
services
pick
sort
of
what
they
put
or
how
they
serialize
those
can
all
be
customized,
but
the
general
concept,
if
it's
the
same
that'll,
still
allow
a
lot
of
code
to
be
shared.
It
should
at
least.
E
And
I
guess
the
I
guess
the
other
question
is:
how
does
how
does
the
the
propagation
for
integration
story
kind
of
fit
with
the
safe
generation
of
x-ray
data
also
fit
into
this
story?
E
Right
because
that's
that's
the
other
kind
of
shard
of
glass
right
that
we
previously
hit
in
other
integrations
that
literally
caused
like
a
chop
review
last
year,
was
not
handling
the
data
properly,
not
handling
threads
properly
and
then
and
then
causing
that
data
leak
between
customers
and
and
how
we
can
expedite
teams
doing
not
only
passive
integration,
but
the
active
integration
of
recording
customer
data
in
a
safe
way.
E
Yeah,
but
I
guess
I
guess
the
thing
is-
is
that
this
kind
of
loops
back
to
the
same
kind
of
conversation
that
we
had
right,
we
can.
We
can
build
an
internal
sdk,
but
we
want
to
have
it
open
source
right.
So
what's
the
gap
there
and
are
we
talking
about
doing
another
new
component
for
teams
to
use
that's
safer
or
are
we
talking
about
fundamental
changes
to
the
existing
tools.
F
With
columbus
also,
but
I
think
we
will
for
the
existing,
I
mean
eventually,
we
want
to
switch
to
open
telemetry.
For
sure
I
mean
the
sdk
is
just
much
better
than
what
we
have,
but
we
don't
want
to
force,
especially
the
current
integrations
to
rewrite
everything,
at
least
in
one
go
and
so
definitely
planning
on
having
a
bridge,
so
we
can
still
use,
for
example,
open
country
based
context,
implementation
with
the
extra
sdk.
F
F
E
Well,
the
the
fact
is:
is
that
trace
just
got
1.0
release
for
various
different
languages
right
are
we
are
we
certain
of
the
safety
that
it's
better
than
better
than
what
we
have
for
the
other
other
x-ray
equivalents
right
now?
E
I
guess
that
that's
the
major
point
right
is
that,
although
yes,
the
the
existing
solutions
are
not
great
and
they
do
have
shards
of
glass
right,
they're
proven
in
the
sense
that
we've
done
integrations
across
probably
a
dozen
teams
now
using
using
the
same
or
similar
tooling,
either
like
shave
down
sdks
kind
of
built
into
little
mini
custom
solutions,
but
that's
that's
kind
of
where
the
the
concern
lies
right.
E
F
Yeah
I
mean
I
don't
know
what
steps
would
be
required
to
convince
like
we
do
have
lots
of
benchmarks
in
general.
I
don't
know
if
we
have
anything
that
compares
x-ray,
stick
and
open
thumb
d
directly,
but
maybe
that
would
be
helpful,
but
I
mean
I
know
like
just
looking
at
the
code.
It's
sort
of
obvious
that
open
thumb
test
is
at
least
as
efficient
and
safe
as
the
extra
sdk
that
I
can
confidently
say
so
just
need
to
figure
out
what
the
best
way
is
to
present
the
data.
For
that
I
think.
D
With
the
context
bridge
that
that
you're
proposing
like
if
we
were
to
backfill
some
of
the
integrations
with
that,
do
you
think
there
would
be
a
way
of
like
doing
like
a
blue
green
deployment
or
something
and
like
comparing
data
like
using
solely
x-ray
integration
versus
like
the
bridged?
One
would
definitely
be
a
good
thing
to
try
it.
I
think.
F
If
integration
is
yeah,
actually
that
so
that
didn't
come
to
mind,
I
forgot
to
mention
this
ever,
but
I
guess
figuring
out
how
to
start
with
x-ray
on
x-ray.
For
this
work
will
be
nice,
like
maybe
the
bridge,
we
should
try
to
roll
it
up
to
x-ray
first
and
then
we
can
get
some
numbers
and
see
how
it
works
there.
Before
going
to
any
of
the
other
teams,.
D
Right
yeah,
I
think,
that'd
be
a
really
good
idea.
They
just
upgraded
well,
not
just
anymore,
but
they
are
using
a
relatively
recent
version
of
the
x-ray
sdk
and
so
yeah
mostly
compatible
with
anything
that
we
build
yeah.
F
E
I
think
we're
trade
treading
in
the
right
direction.
Obviously,
nozzles
are
super
keen
on
on
on
unseeing
the
dock
and
and
having
this
discussion,
as
I
said,
dennis
and
nazar
are
viewing
this
as
like
two
separate
things,
one
to
support
w3c
and
one
to
like
improve
our
integrations.
E
So
so
it
might
be
helpful
to
kind
of
sort
of
write
the
doc
in
that
retro
in
that
perspective,
make
that
separation,
but
at
the
same
time
we
can
still
pitch
it
as
like.
One
thing,
because
I
I
agree
with
you
like
integrations
is
propagation.
It
is
easy
onboarding,
it
is
handling
customers
data
safely
and
in
a
safe
way.
E
Right
like
you
like,
I
see,
I
see
propagation
and
integrations
much
more
intertwined
than
than
than
the
the
other
audience
but
yeah,
I
guess,
are
there
any
action
items
or
next
steps
or
things
that
we
should
be
thinking
about,
and
when
do
you
think
we
can
have
a
first
pass
of
the
dock,
at
least
for
discussion
purposes
with
with
dennis
and
nazar.
F
Yeah,
like
I'm
hoping
to
finish
today,
I
mean
this
week.
The
doc
will
be
done,
so
we
can
review
it
next
week.
I
think.
E
Okay
cool:
I
didn't
realize
that
you're
that
confident
so
fast,
okay,
cool
at
least
for
her
first
draft
anyways
right,
yeah,
cool
yeah
paul
well
feel
free
to
throw
it
around
to
us.
If
you
want
some
comments
before
we
take
it
to
the
audience
alex
william,
how
are
you
feeling?
Are
there
any
comments,
concerns
or
call
outs
or
things
that
we
we
need?
We
should
be
thinking.
A
So
just
a
little
bit
of
history,
like
definitely
x-ray
integrations,
had
its
own
scope
and
then
in
parallel,
lambda
cloudwatch,
logs
and
s3
we're
looking
at
some
joint
effort
how
to
detect
usage
loops
and
prevent
them
from
happening
because
customers
just
write
some,
you
know
lambda
triggered
by
s3,
then
again
puts
in
same
bucket.
A
A
A
So
one
of
the
objectives
of
all
this
exercise
is
to
see
first
of
all,
whether
like
helping
and
solving
this
lineage
or
like
detecting
into
that
loop
aligns
with
our
long-term
vision
in
x-ray,
and
another
objective
is:
how
do
we
come
up
with
a
solution
that
requires
minimal
effort,
especially
repeatable
efforts
on
the
service
teams
like
s3
to
propagate
these
contacts
in
that
context,
etc?
A
So
I
did
a
like
an
exercise.
A
Over
the
weekend,
I
actually
implemented
infinite
loop
and
enabled
active
tracing
in
those
lambdas,
and
if
you
look
at
the
service
side
like
what
we
do
on,
you
know
back
in
exactly
perfectly
looking
as
a
yeah,
you
can
totally
detect
this
usage.
Look
like
say
it
shows
it
right
there.
A
As
a
matter
of
fact,
the
problem
is
similar
to
our
frozen
trees
situation
when
same
trade
just
keep
being
updated
with
more
and
more
segments,
and
we
decide.
Oh
it's
just
too
large.
Let's
stop
doing
it
so
like
we're
kind
of
really
doing
something
similar,
so
xa
could
potentially
take
on
that
role
of
you
know:
detecting
trace
with
repeatable
segments
and
sending,
like,
I
don't
know,
event
to
advantage,
etc.
A
So
this
and
the
next
question
I
think,
then
becomes
okay.
If
we
ask
x-ray
server-side
to
solve
it,
what
else
do
we
need?
Does
it
align
with
our
long-term
vision?
A
And
if
you
look
at
like
what
requirements
for
propagating
lineage
are
some
of
them
actually
in
line
with
our
long-term
vision,
for
example,
for
lineage?
What's
critical?
Is
that
how
would
that
context
be
propagated
with
zero
customer
coding
instrumentation,
because
and
so
we're
talking
to
the
adwords
sdk
team?
Saying
hey?
Can
you
propagate
this
lineage,
but
same
time?
It's
like
I'm.
Listening
to
this
and
thinking
wait
a
minute
like
I'd
like
a
double
sdk
to
propagate
trace
header.
A
Why
would
they
only
propagate
lineage,
and
so
in
that
aspect,
I
think
there
is
a
lot
of
like
you
know
like
this
word,
like
synergy
right
so
between
the
two,
and
there
is
a
lot
of
attention
to
solve
that
usage,
loops
problem
and
brian
wants
us
to
at
least
present
an
option
where
x-ray
is
spreading
solution
for
usage,
loop
detection
and
we
build
ourselves
in
into
this
effort
to
get
visibility,
hopefully
get
more
funding,
but
same
time.
A
Requires
zero
instrumentation
for
the
customers
so
yeah.
We
can
then
discuss
what
that
basic
tracing
means,
how
we
accept
the
cost,
how
we
make
it
affordable,
but
ultimately,
like
then
the
first
step
is:
how
do
we
propagate
the
context
across
aws,
sdk
and
all
the
services?
Then
the
second
one
that
will
already
benefit
essentially
x-ray.
A
As
we
build
this
network,
then
we
then
go
ahead
and
you
know
just
extend
our
passive
integrations
into
active
integrations
and
so
when
those
services
emit
segments
when
customer
code
hits
them
with
propagated
headers
that
it's
like
on
you're.
Looking
at
the
picture
that
is
developing
at
the
moment,
so
all
these
nodes
will
start
appearing
on
this
picture.
As
we
add
more
integrations.
A
D
A
Indicator
no,
not
really
like
I
actually
was
trying
to.
I
was
trying
to
explain
the
mode
when
we
don't
need
any
baggage
at
all
right
like
imagine.
If
we
we
could
solve
what
I
was
trying
to
explain.
We
could
solve
this
purely
on
the
server
side.
D
A
But
it
doesn't
right
so
so
this
is
the
lineage
approach.
When
x-ray
does
not
hold
the
state,
the
state
is
captured
inside
the
lineage
and
how
we
propagate
lineage
is
another
problem,
but
like
lineage,
is
the
state
or
visitation
right,
but
I
think
there
is
a
world
when
we
solve
everything
on
the
server
side
by
x-ray,
back-end
and
the
question
is
like
what
how
would
benefit
from
it,
and
I
think
the
benefits
are
actually
really
substantial.
A
But
what
made
so,
I
think,
like
just
to
summarize
three
options-
are
considered
one:
that's
still
an
option
on
the
table.
What
if
we
propagate
more
user
data
on
the
existing
xml
trace
id
header,
that
user
data
will
represent
image?
Another
approach
is
just
like
lineage
is
its
own.
Header
does
not
interfere
with
tracing
in
any
way,
except
that
we
may
have
a
universal
way
of
propagating
that,
which
is
what
enron
was
describing.
E
Okay,
thank
you.
Thank
you,
so
much
yeah
this
is
this
is
great.
Honorable
will
alex
thanks
for
thanks
for
attending
on
short
notice,
and
then
we
can
set
up
a
slack
channel
if
that's
helpful
at
all,
or
we
can
just
chat
offline
or
use
this
weekly
sync.
D
D
D
D
K
K
B
B
K
B
F
B
E
E
No,
no,
let's,
let's
chat
now,
it's
already
all
in
my
brain
anyways
I've
been
reviewing
his
document
with
him
over
the
past
two
weeks,
so
I've
got
a
good
idea
of
everything,
that's
in
there
and
going
over
all
the
role,
guidelines
and
stuff.
So
it's
your
first
review.
I
guess
formal
feedback
review,
so
I'm
here
to
kind
of
help
you
through
understanding
that
whole
process
and
everything
so
so,
stop
me
if
I've
done
this
before
so
I
sent
you
a
couple
of
things.
I
sent
you
the
sd
roll
guidelines.
E
I
sent
you
the
work
history
document
that
we
were
working
on
with
them
and
then
I
think
I
also
sent
you
my
101
doc,
but
that's
not.
It
doesn't
really
need
to.
E
Yeah,
that's!
That's!
Fine!
That's
more
of
a!
I
feel
like
it's
almost
like
stretch
goals
for
for
sds
in
general.
It
doesn't
really
fit
the
quote-unquote
role
guideline
criteria
per
se,
maybe
in
a
couple
of
points
later
on,
but
we
can
probably
review
that
at
a
later
time
but
yeah.
So
you
had
a
chance
to
read
over
the
roll
guidelines.
A
F
E
Yep
same
thing:
yep,
okay,
I
didn't
realize
you
already
done
one
before
okay,
perfect.
F
Yeah,
it
was
a
I
mean,
not
for
an
engineer
that
was
for
so
yeah.
I
think
one
of
the
ones
that
worked
with
us
for
open
telemetry,
and
so
I
think
they
just
oh
any
senior
engineer
you
can
find
you
get
some
feedback
and
that's
a
bit
of
light
feedback.
Yeah
yeah.
E
That's
that's
that's.
The
general
thing
like
I
reached
out
to
norm
norm
is
never
actually
actively
worked
on
any
of
our
sdks
or
or
exactly
with,
like
any
member
of
the
team,
but
you
know
he's
a
senior
over
there
and
he's
been
involved.
So
I
had
him
review
the
work
history
and
give
feedback
as
well.
So
you
got
to
do
what
you
got
to
do.
E
F
Exactly
all
right
so
yeah,
I
think
I'm
okay
on
background.
I
guess
the
one
just
reading
through
prashant's
talk.
Most
of
it
made
sense.
The
one
thing
that
sort
of
stood
out
was
of
them
like,
for
example,
the
benchmarking
is
specifically
in
my
court
also,
but
those
that
was
such
a
small
task
like
is
that
something
worth
putting
into
a
promo
doc
is
sort
of
one
thing.
I
one
question
I
had,
I
guess
yeah.
E
No,
I
I
I
removed
that
I've
been
kind
of
I've
been
helping
them
put
this
because
the
promo
doc
is
short.
It's
only
1500
words
right,
so
basically
what
I
had
them
do
with
the
work
history
document
is
list
everything
that
he's
done
everything
and
then
we
just
cut
it
down,
got
it.
Okay,.
F
K
K
E
Promotion,
doc
in
order
to
speak
to
the
feedback,
so
it's
actually
a
reverse
process,
you're
supposed
to
get
the
feedback
first
and
then
write
the
document
around
it,
yeah,
okay,
because
because
they're
really
going
to
drill
down
on
the
reasons
not
to
promote,
which
is
fine-
and
that's
that's
great,
if
you
have
really
strong
reasons
not
to
promote
that's
more
things
that
I
can
kind
of
anchor
the
dock
in
in
regards
to
to
either
covering
that
and
saying
that
this
is
something
that
we're
working
on.
E
This
is
something
that
has
improved
or
his
other
strengths
kind
of
outweigh
this,
even
if
it's
an
area
of
weakness,
but
this
is
what
I've
had
planned
forward.
So
everyone
everyone's
going
to
have
weaknesses
right,
no
one's
completely
perfect!
You
know
no
one's
going
to
hit
all
the
bullet
points
across
the
board.
That's
completely
fine!
I
just
have
to
make
sure
whatever
feedback
that
you
have
is
accurately
displayed
in
the
document,
and
there
are
some
kind
of
steps
forward
from.
F
H
F
As
well
so
tell.net-
and
even
just
I
mean
I've
talked
to
like,
I
know
his
general
work
also,
so
I
can
comment
on
that.
Maybe
the
only
specific
project
in
this
stock
I
can
mention
is
the
hotel.net
how
he
drove
through
the
country,
repo
stuff,
which
was
pretty
nice.
E
Yes,
the
other
one:
that's
in
there
is
the
asset
pipeline.
E
Yeah,
it's
it's
gonna,
be
a
challenge
for
anybody
to
have
his
full
background,
because
we've
had
that
change
of
teams
and
management
and
co-workers
so
often,
so
I
am
kind
of
scatter
shotting
it,
but
at
the
same
time
this
is
pretty
standard
practice
not
for
lford
l5
but
l5
to
l6.
You
have
what's
called
a
assessment
advisor,
so
they
actually
go
through
it's
a
third
party
person
that
they
bring
in
for
the
review.
E
They
review
the
promo
document,
they
review
the
assets,
they
review
the
deliverables
and
they
review
the
code
and
they
say
yes,
this
this
person
meets
all
the
criteria
for
l6
or
not
so
I'm
kind
of
basing
it
on
that
too,
whereas
you
guys
are
in
in
trusted
spots
right,
you're,
very
technically
qualified
for
your
position,
obviously
you're
a
senior
engineer
on
the
open
source
team,
so
I
can
give
you
artifacts
and
you
can
backfill
from
that.
F
E
The
way
that
dennis
likes
the
feedback
written,
it
should
call
out
lps
both
for
strengths
and
weaknesses
and
then
call
out
explicit
examples
if
you
have
any
for
strengths
and
weaknesses
as
well,
because
we're
very
again
data-driven.
We
should
have
concrete
examples
before
making
generalized
statements.
At
least
one
right.
You
can
say:
oh
prashant
failure
fails
to
deliver
because
xyz
right,
if
you
have
the
data
points
for
it,
that's
fine.
E
You
can
put
it
in
there
and
I
can
say,
oh
well,
he
may
not
have
delivered
in
this
case,
but
he
delivered
in
these
cases
and
the
reason
why
he
didn't
deliver
in
this
case
was
because
xyz.
This
is
what
we're
working
on
in
the
future
and
how
we're
going
to
get
better
at
it,
and
he
showed
improvement
with
this
future
project
or
this
current
initiative
or
something
like
that.
So
again,
completely
fine,
very
normal.
You
can.
You
can
be
very
harsh
on
on
good
feedback.
E
F
Yeah
yeah
like
it
was
funny
because
when
I
did
the
other
promo
review
thing,
I
sent
one
reply
where
I
had
very
little
stuff
in
the
second
box,
because
at
google
generally
you
don't
put
too
much
information
about
weaknesses
for
promo
for
whatever
reason,
but
I'm
glad
to
hear
that
here
it's
more
thorough
like
there's
actually
one
time
where
I
wasn't
in
favor
of
promotion
for
someone,
and
they
asked
me
if
I
should
write
feedback
because,
like
I
guess
so,
I'm
you're
supposed
to
collect
all
the
feedback
you're
supposed
to
get.
F
E
And
yeah
it
doesn't
even
need
to
be
like
all
supportive
either
if,
if,
if
it's
a
couple
of
people
on
there,
that
says
say
that
they
don't
support
and
here's
the
reasons
why
I
just
argue
in
my
case
and
say
like
that
might
be
the
case.
But
you
know
here's
the
here's,
my
reasons
why
I'm
backing
this
yeah,
so
cool
yeah,
yeah.
F
E
Yeah,
actually,
I've
so
will's
actually
really
good
at
writing.
I
don't
and
he
wrote
his
own
promotion
document
as
well.
That's
usually
how
I
go
about
it
with
engineers.
It's
really
really
good.
I
really
don't
agree
with
the
whole
manager
should
write
the
promotion
document
because,
while
you're
writing
the
promotional
document,
you're
forced
to
reflect,
you
have
to
think
of
all
these
things
that
you
did.
You
know
what
what
could
have
been
improved
on.
What
didn't
go
right
right,
it's
it's
effectively!
E
Your
own
feedback
mechanism
is
keeping
yourself
honest
and
going
through
your
work,
history
right.
So
it's
it's!
It's
been
a
bit
of
a
challenge
because
sean's
not
as
great
as
a
writer,
so
I've
been
having
to
coach
him
on
like
using
biased
wording
and
unbiased
wording.
So
is
his
his
original
draft
that
he
sent
me
was
like
diligence
took
like
energetic
approaches
like
that's,
not
data
driven
yeah
here.
Let
me
grab
this
because
again,
like
you're
great
at
writing
too,
but.
E
This
is
the
feedback
that
actually
no
I'll
send
it
to
you
after.
E
I
don't
yeah
because
I
can't
can't
taint
your
original
feedback
right.
Okay,
that's
a
bad
idea!
I
was
going
to
send
you
an
example,
but
no
I'll
I'll,
just
I'll
just
ask
questions
on
your
feedback.
If
anything
needs
to
change.
Okay,
cool!
That's
pretty
much!
All
I
wanted
to
chat
about.
If
you
have
questions
on
the
weird
ambiguity
and
the
role
guidelines,
I'm
happy
to
help
chat
about
that.
I
know
it's
really
kind
of
stupid.
E
K
E
I'm
targeting
for
the
document
to
be
done
by
the
20th.
So
I
know
that
doesn't
give
too
much
time
but
at
the
same
time,.
F
F
F
If
it's
weird
or
you
don't
know
just
say
it,
but
if
I'm
wondering
if
you
know
anything
about
fertility
clinics
in
seattle,
the
reason
being
that
thinking
about
having
a
kid
and
I
have
sort
of
a
2.5
genetic
disease
and
don't
want
to
pass
it
on,
but
they
don't
do
that
in
japan.
They
only
do
so
thinking
about
maybe
spending
some
time
in
seattle
for
that
and
want
to
find
a
good
place,
and
I
don't
know
how
to
do
that.
F
E
Let
me
let
me
go
ask
you
know
what
there's
so
there's
awe,
which
is
amazon
women
in
engineering.
I
bet
they
have
no
idea
about.
F
F
E
Medical
stuff,
like
oh
yeah,
completely
unrelated
as
well
lena,
has
her
trachea
shave
booked
so
she's
going
under
the
knife
in
another
couple
of
months.
So
she's
super
excited:
okay,
nice,
yeah,
yeah,.
E
Yeah
yeah,
we
got
a
really
really
nice
surgeon
he's,
like
top
rated
in
washington
state
for
like
the
past
five
years
yeah.
So
he
does
this
every
like
several
times
a
month,
so
she's
super
excited,
but
we
met
him
for
the
consultation
last
week,
which
is
out
of
work,
so
just
yeah
she's,
just
just
like
totally
comfortable
with
them
and
she's
like
yeah
this.
This
is
the
right
guy.
I'm
like
okay,
let's.
A
F
Yeah
yeah
so
the
genetic
testing,
so
it's
the
genetic,
it's
called
pgt
and
it's
also
not
that
easy
of
a
procedure
which
is
sort
of
why
I'm
caring
more
like
that
they're.
Actually
I
mean
I
don't
think
they
have
to
be
the
top
in
the
world
or
anything,
but
as
long
as
they
have
a
reasonable
record
of
successes
or
something
is
sort
of,
I
don't
know
how
to
find
that
information.
E
Yeah
yeah
absolutely
yeah,
as
I
said,
I'll
go
digging
around
and
stuff
and
stuff,
not
a
problem.
E
Sounds
great
bug
me
anytime,
we
should
probably
meet
for
more
often
one-on-ones,
because
yeah.
E
Yeah,
whatever
kind
of
works
for
you,
I
know
the
schedules
are
all
kind
of
weird,
but
but
yeah
just
throw
it
down.
You
can
even
do
monday
evenings
if
you
want
to
whenever
your
availability
is
good.
Okay,
cool
hi
yeah
have
a
great.