►
From YouTube: Humans of DataHub - Fredrik Sannholm
Description
This week we are excited to share our second installment of Humans of DataHub. We are joined by Fredrik Sannholm of Wolt, a Finnish tech company known for its delivery platform for food and other goods.
We had the pleasure of chatting on Zoom, check out our conversation to hear about what brought Fredrik’s team to DataHub, what DataHub has enabled for Wolt’s organization, and so much more!
A
All
right
well
welcome
to
humans
of
data
hub
we're
going
to
kick
things
off
with
an
introduction
from
frederick.
So
frederick
tell
us
a
little
bit
about
yourself.
B
Yes,
hello,
my
name
is
frederic
sunholm.
I
work
as
a
engineering
team
lead
at
a
company
called
vault,
which
is
not
very
unlike
your
your
door
dashes
and
deliveroos
and
ubereats
of
the
world
we
operate
in
in
in
europe,
asia
and
and-
and
so
maybe
the
american
viewers
haven't
haven't,
maybe
heard
of
us
so
much
but
we're
big
in
big
in
japan.
I
lead
the
team.
I'm
elite
is
a
sort
of
data
in
data
engineering
and
machine
learning,
engineering,
sort
of
infrastructure
team.
B
So
so
we
handle
all
of
our
our
infrastructure
on
on
both
of
those
sites,
pretty
small
team,
but
we
we
consist
of
like
a
couple
of
data
engineers,
a
couple
of
machine
learning
engineers
and
some
devops
engineers,
and
maybe
tells
us,
tells
you
a
bit
about
the
like
focus
of
of
the
team,
very,
very
sort
of
infra
heavy.
A
Awesome,
and
when
did
you
so
I
know
that
I
know
that
you've
been
in
the
data
hub
community
for
a
while
when,
like
how
did
you
first
learn
about
data
hub?
What
did
you
join?
What
brought
you
to
us.
B
C
B
B
Of
course
things
have
changed
now
and
there's
a
lot
of
different
options,
but
but
we
we
sort
of
yeah
just
stumble
upon
a
lot
of
proprietary
solutions
that
we
weren't
very
interested
in
and
and
then
one
couple
of
open
source
ones,
but
they
didn't
support
lineage,
and
then
we
actually
found
one
this
one
very
sort
of
a
less
modern
that
we
we
sort
of
it
ticked
all
our
boxes
at
the
time
and
we
we
sort
of
yeah.
C
B
Try
it
and-
and
I
I
beg
my
my
my
colleague
or
team,
lead
for
the
platform
platform
team
and
got
a
devops
engineer
to
help
me
out
to
deploy
it
and
probably
spent
like
a
month
on
that
on
that
thing
and
and
got
absolutely
nowhere.
B
And
this
devops
engineer,
I
just
like
sensed
it
from
his
slack
messages
that
he
he
was
like
losing
losing
his
nerve
on
me
and-
and
I
just
had
to
find
another
solution,
and
then
I
don't
know
how
I
miss
it
the
first
time
around,
but
stumble
upon
like
data
hub
and
at
this
point
it
was
linkedin,
was
primarily
developing
it
still
and
it
sort
of
ticked
most
of
our
boxes,
not
maybe
all
of
them,
but
just
the
fact
that
they
had
like
up-to-date
docks.
B
You
know
you
could
deploy
it
locally
with
one
command
yeah
it
had
lineage
tracking,
so
so
that
was
big
plus
and
it
was
sort
of
modern
technology.
Kafka
kubernetes,
all
that
stuff.
So
so
we
just
like,
went
for
it
and
it's
been
a
success
story
ever
since
and
yeah
so
good
thing
is
that
that
we're
still,
in
speaking
terms
with
this
devops
engineer
so.
B
It's
it's
really
extraordinary
and
I
have
to
be
honest:
I'm
not
I'm
not
trying
to
like
sell
this
to
anyone,
but
it's
it's
like
really
one
of
a
kind
in
my
opinion,
but
we
are
sort
of
avid
user
of
different
open
source
solutions.
We
do
try
to
engage
in
with
the
community
as
much
as
possible
just
to
to
well.
Sometimes
you
just
need
help
with
some
problem
solving.
C
B
Just
get
a
get
a
feel
for
for
like
or
get
get
in
contact
with
the
core
developer
team
and-
and
I
think
data
hub
is
the
github
community
or
just
like
slack
channel.
There
is
it's
just
just
super
super
like
unique
and
and
everyone's
so
super
helpful.
You
know
you
get.
A
B
B
It's
like
spot
on
the
troubleshooting
guides
and
everything.
So
it's
just
like
the
responsiveness
basically
and
the
like
helpfulness
of
the
whole
whole
community
and,
of
course,
all
the
all
the
like
great
great
contributions,
especially
on
the
like
ingestion
side,
popping
up,
left
and
right,
different,
different
or
support
for
different
sorts
of
ingestion
sources.
So
that's
really
really.
A
C
A
B
Maybe
first
and
foremost
has
made
our
our
legal
team
very
happy.
So
that's
a
big
bin
for
everyone.
They
they
they're
very
happy
with
just
being
able
to
to
keep
track
of
of.
Where
is
our
sensitive
data?
Where
is
it
going
how's
it
being
processed?
Where
is
it
ending
up.
A
B
Sending
it
to
you
know
which
third-party
tool
third-party
apis
and
especially,
why
are
we
sending
it
there
and
and.
A
C
B
Displays
in
very
sort
of
easily
searchable
format
and
clear
text,
explanations
of
on
all
the
columns
in
our
snowflake
snowflake
warehouse,
it's
just
great
for
onboarding
and
and
great
for
yeah
new
people
joining
and
not
being
very
familiar
with
with
with
with
the
warehouse
or
the
data
in
general.
So
that
helps
us
a
lot
and
then
we've
actually
implemented
a
couple
of
a
couple
of
cool
cultural
technical
features
around
it
as
well
like
related
to
like
data
quality
monitoring.
So
so,
oh.
A
B
So
you
know
we
can
keep
a
look
at
like
the
the
quality
in
a
specific
table,
and
then
data
hub
has
all
the
metadata
that
we
need
like
who's
the
owner.
Where
is
it
coming
from.
B
You
know,
airflow
job
is
processing
it
and
things
like
this
and
and
what
is
expecting,
who
is
it
affecting
downstream
yeah?
Maybe
stop
like
a
machine
learning
job
because
of
the
data
being
stale,
for
example,.
C
A
Yeah
awesome,
that's.
It
makes
me
really
happy
to
know
that
this
is
happening
in
practice,
because
you
know
they're
just
selfish.
The
reason
I
joined
the
data
hub
community
and
joined
this
team
is
because
I
felt
the
pain
of
doing
all
that
manual
troubleshooting
when
something
when
something
failed,
and
so
it
just
makes
me
happy
to
know
that
this
doesn't
work.
C
B
Yeah
I
mean,
as
as
you
might
have
figured
already
we're
very
keen
on
this.
This
lineage
aspect
of
the
of
the
products
and
actually
my
eye
on
on
one
particular
feature.
That's
I
think
it's
in
progress
at
the
moment,
so
so
this
possibility
to
to
like
more
flexibly,
query
like
the
other
downstream
entities
and
all
the
metadata
about
them.
B
That's
one
thing
that
we
had
to
work
around
a
bit
in
in
some
of
our
some
more
solutions
and-
and
we
we're
really
really
hoping
for
for
like
a
native
native
solution
for
that.
So
so
so
that's
one
where
there's
a
practical
practical
thing
that
that
we're
looking
for
we're
big
snowflake
snowflake
users,
so
everything
we're
very
happy
to
see
all
the
sort
of
new
feature
or
metadata
collection
features
that
that
you're
bringing
in
usage
and
data
quality
and
all
this
stuff-
that's
probably
in
there
already.
B
But
but
we
haven't
really
had
the
time
to.
B
A
So
we
can
I'm
excited
for
all
of
those
things
too,
and
you
should
come
to
town
hall
next
friday
because
we'll
be
demoing
the
downstream
impact
awesome
stuff,
which
is
super,
exciting,
yeah.
Anything
else
like
any
like.
What
would
you
do?
You
have
any
advice
for
new
people
that
are
new
to
data
hub
either
getting
started
with
it
getting
familiar
with
it,
make
it
successful
within
the
organization.
B
I
I
think
the
only
only
only
advice
I've
just
joined
the
slack
channel
day,
one
it's
just
just
a
great
source
of
of
knowledge
and
tips
and
helpful
people,
and
I
mean
the
documentation-
is
already
great
but
there's
just
so
much
new
stuff
being
added.
B
It's
difficult
to
just
keep,
checking
the
documentation
all
the
time.
So
so,
just
like
listening
in
or
yeah
joining
the
announcement
channel,
basically
and.
B
A
B
C
B
It's
gonna
be
the
announcements
one.
It's
the
one.
C
C
A
B
And
then
every
time
I
miss
a
town
hall,
I
just
I
know
the
the
link
to
the
the
recording
is
there.
So
I
just
yeah.