SPEAKER_00
0:42
In
my
experience,
I
can't
speak
out
for
everybody,
but
I
often
find
that
the
pushback
isn't
about
the
technology
itself.
The
pushback
is
about
the
trust
and
the
uncertainty
of
what
it's
going
to
do
for
the
roles
that
essentially
are
getting
asked
to
use
this
technology.
Some
of
the
experiences
just
in
the
time
that
I've
been
with
some
of
our
customers
is
really
figuring
out
is
this
going
to
create
more
work,
right?
Is
it,
can
I
trust
it?
Is
it
going
to
know
what
to
do?
There's
a
lot
of
very
valid
questions.
When
these
questions
surface
at
the
start
of
a
project,
let's
say
fast
forward
two
weeks,
three
weeks,
and
you
see
the
smiles
on
their
face.
You
see
the
aha
moment
of,
hey,
you
know,
you
told
me,
but
I
really
wasn't
sure.
As
a
product
leader,
that's
probably
one
of
the
most
exciting
parts
of
it.
SPEAKER_02
1:35
Welcome
in
to
our
Cargo
Rex
series
on
AI
use
cases
in
logistics.
We've
got
another
great
conversation
for
you
today.
And
we
are
talking
with
Michelle
McBride.
She
is
the
head
of
product
over
at
Envoy
AI,
where
they
are
helping
logistics
service
providers
not
just
look
at
dashboards
with
a
bunch
of
data,
but
take
action
on
that
data.
So,
Michelle,
excited
to
have
this
conversation.
SPEAKER_00
1:57
I
am
so
happy
to
be
here.
Thank
you
for
having
me.
SPEAKER_02
2:00
Absolutely.
Now,
now
for
we
were
talking
before
we
hit
record
on
your
extensive
experience
within
the
industry.
16
years
at
Uber
Freight.
And
then
after
16
years,
what
made
you
take
the
leap
over
to
Envoy?
SPEAKER_00
2:15
Great
question.
So,
yeah,
you
know,
spend
16
years
in
transportation
management,
implementation,
solution
designing,
all
of
the
above
you
can
think
of.
And,
you
know,
the
opportunity
where
we're
sitting
at
right
now
as
it
relates
to
what
the
industry
is
shifting
to,
the
transition
of
how
to
perform
at
a
faster
scale,
how
to
get
fast,
how
to
get
more
intelligent
with
the
way
that
you
do
operations
really
excited
me.
So
when
I
had
an
opportunity
to
meet
Robert
Mason
at
Envoy,
it
was
just,
it
was
awesome.
It
got
me
excited.
The
vision,
but
also
the
problems
that
we're
looking
to
solve
through
the
usage
of
this
tag
was
really,
really
exciting
for
me.
And
yeah,
here
I
am.
I'm
currently
responsible
for
product
strategy,
uh,
the
product
roadmap
behind
our
application
Ling,
which
we'll
get
into
here
in
just
a
minute.
Um,
but
yeah,
our
focus
is
just
to
ensure
that
we're
able
to
reorchestrate
the
way
operations
get
done.
And
so
let's
talk
about
from
a
high
level,
what
does
Envoy
do?
Yeah,
absolutely.
So
we
are
think
of
think
of
Envoy
as
the
orchestration
layer,
the
intelligent
orchestration
layer
for
logistics.
When
you
look
at
the
workflow
of
a
carrier
cell
screp
or
somebody
sitting
out
on
the
logistics
floor,
some
of
the
most
common
challenges
are
how
do
I
get
this
load
booked?
How
do
I
negotiate
it?
How
do
I
price
it?
How
do
I
make
sure
that
I
am
using
the
right
carrier?
How
do
I
keep
up
with
exceptions?
And
I
think
traditionally,
having
been
in
the
industry
for
16
years,
we
have
encountered
an
evolution
of
how
those
tests
got
done,
right?
In
the
past,
early
times
when
I
was
in
operations,
it
would
be
phone
call,
it
would
be
email.
Uh,
then
we
started
to
get
into
the
era
of
EDI
and
APIs.
And
we
started
to
see
more
digitization
on
how
we
communicated
with
carriers
or
other
providers
in
the
industry.
Here
with
Envoy,
as
you
think
about
it,
we
are
essentially
orchestrating
a
multiple
uh
tech
stacks
or
multiple
systems
that
users
have
already
today,
and
we're
essentially
adding
intelligence
on
top
of
it
to
help
execute
some
of
these
tasks
in
a
much
more
automated
fashion.
So
it's
really
looking
at
it
from
an
order
to
cash
perspective
and
what
are
the
various
workflows
that
LA
taps
into.
But
yeah,
ultimately
what
we're
trying
to
establish
and
trying
to
fix
is
the
operational,
but
taking
away
the
operational
pain
in
the
operational
burden
into
more
of
a
scalable
AI-driven
workflow
for
operations.
SPEAKER_02
4:45
And
so
is
that
operations
inside
of
a
brokerage,
inside
of
a
you
know
a
large
carrier?
What
does
that,
I
guess,
sort
of
customer
makeup
look
like?
SPEAKER_00
4:53
Great
question.
So
right
now,
our
current
focus
has
been
with
the
with
the
brokerage
industry,
um,
really
providing
support
to
or
providing
this
technology
to
the
carrier
sales
rep.
So
those
that
are
focusing
on
getting
the
loads
covered,
getting
the
freight
moved,
the
brokerage,
you
know,
managers,
supervisors
who
are
really
looking
for
intelligence
to
operate
their
business
better,
right?
So
it
it
levels
up
to
through
our
analytics
and
through
different
layers
of
information
at
that
level.
But
if
you
think
about
it,
this
technology
also
can
apply
across
other
verticals
as
well.
Companies
that
are
running
automated
business
processes
or
BPOs
type
type
of
environments
where
they
are
currently
sitting
with
repetition
and
task.
The
only
way
to
scale
is
to
add
additional
labor
into
their
into
their
operation.
This
technology
essentially
has
the
ability
to
get
embedded
in
the
ecosystems
that
they're
operating
on
today
and
help
them
execute
at
a
faster
and
better
scale.
SPEAKER_02
5:52
I
want
to
go
back
to
the
the
carrier
sales
role
because
I'm
curious
what
are
some
of
those
maybe
uh
workflows
or
processes
that
they
are
dealing
with
today
and
and
where
are
they
kind
of
struggling
where
Ellie
fits
in?
SPEAKER_00
6:07
Yeah,
a
great
question.
So
if
you
think
about
it,
let's
first
put
our
hat
on
of
the
carrier
rep.
How
do
we
reduce
the
operational
burden,
the
drag,
right?
Finding
carriers,
verifying
compliance
for
those
carriers,
managing
the
outreach.
We
know
today
that
not
everybody's
at
the
same
level.
Some
carriers
in
terms
of
outreach
are
going
to
be
email-based.
Others
to
this
day
and
age
are
also
phone
calls.
Then
you
have
the
ones
that
are
leveraging
WhatsApps
or
other
um
methods
of
communication.
So
managing
the
outreach,
the
management
or
the
handling
of
repetitive
follow-ups.
Hey,
are
you
able
to
pick
up?
Hey,
did
you
arrive?
Hey,
what's
your
ETA?
Things
that
are
just
fundamentally
critical
to
the
execution
of
freight
that
happen
all
the
time.
There's
no
there's
no
secret
sauce
on
this,
but
it's
just
really
the
method
of
how
we
empower
and
how
we
up-level
through
the
usage
of
products
like
Ellie
to
get
that,
get
those
functions
executed.
It's
essentially
removing
that
friction.
And
then
at
the
same
time,
through
these
stages
that
I
just
described,
you
run
into
the
um
into
a
situation
where
many
of
these
reps
have
five
to
six
to
seven
different
screens
where
they're
having
to
do
these
functions
across
a
group
of
multiple
uh
tools
that
they
already
exist.
Ellie
brings
all
that
together.
SPEAKER_02
7:31
And
so
what
does
that
look
like?
Is
it
maybe
like
a
dashboard
that
you
manage?
Is
it
you
know
an
integration
into
their
existing
TMS?
How
do
you
how
do
you
interact
with
Ellie?
SPEAKER_00
7:41
Yeah,
so
think
of
it
as
a
browser-based
extension.
So
what
it
does
is
the
way
that
it
operates
is
essentially
our
users
will
have
an
extension
that
is
downloaded
into
Chrome
or
to
wherever
browser
of
their
choice.
And
the
back
end
of
Ellie
essentially
is
queued
up
to
be
able
to
interface
with
the
TMS.
It's
able
to
integrate
and
interface
with
compliance
software,
some
of
the,
to
name
a
few,
some
of
the
ones
in
the
market,
highway,
my
carrier
packets
with
from
Descartes,
from
a
posting
and
um
and
negotiation
layer.
You
also
have
integrations
with
companies
like
DAT.
Um,
so
yeah,
through
this
extension,
essentially
it's
connecting
an
ecosystem
and
creating
this
infrastructure
where
the
carrier
rep
doesn't
have
to
take
advantage
or
use
all
seven
screens
to
do
and
essentially
perform
each
of
these
steps.
SPEAKER_02
8:33
And
so
it's
really
bringing
in
together
all
of
those
different
because
I
I
think
if
I
take
a
step
back
a
little
bit
for
a
lot
of
brokerages,
there's
been
just
an
onslaught
of
tech,
you
know,
being
introduced
to
different
brokerages
and
teams
over
the
last
decade.
And
maybe
some
have
overinvested
in
in
certain
areas,
and
now
they're
trying
to
figure
out
how
to
make
sense
of
it
all.
And
that's
what
really
drew
me
to
the
you
know,
the
opening
line
of
how
to
not
just
deal
with
data
and
dashboards,
but
how
to
take
action.
So
once
you're
you're
you're
in
the
the
system
with
Ellie
and
it's
tell
it's
getting
rid
of
some
of
the
tabs.
What
I
guess
how
are
you
interacting
with
the
platform
outside
of
that?
Or
is
it
streamlining
that
part
of
the
process?
And
then
it
kind
of
frees
you
up
to
have
you
know
more
direct
conversations
or
source,
you
know,
better
carriers.
Uh
what
I
guess
are
maybe
are
the
perks.
SPEAKER_00
9:28
Yeah,
great
question.
And
and
I
I
want
to
start
first
with
something
that
you
just
stated,
because
I
spoke
about
it
earlier
about
the
transformation
that
we've
been
through
in
in
freight
operations,
right?
First,
you're
spot
on
with
the
the
gaps
were
how
do
I
get
data,
right?
Where
where
do
I
source
it?
We
all
experience
the
area
of
digitization.
And
you
know,
now
we're
sitting
absolutely
like
you
just
mentioned,
what
do
I
do
with
it?
How
do
I
execute
better?
How
do
I
take
away
so
much
fragmentation?
Tooling,
accessibility
to
tooling
today
is
not
a
problem,
is
not
the
problem.
Many
great
products
out
there,
great
options
for
brokerage
reps
to
be
using
um,
you
know,
at
any
other
level
to
be
taking
advantage
of,
but
it's
how
you
go
about
the
execution
of
that
sets
of
tools
that
you
have
that
are
really
going
to
make
you
become
more
efficient
in
your
daily
operations.
So
when
you
think
about
it
with
Ellie,
the
idea
is
that
by
being
embedded
into
the
workflow
and
removing
the
fragmentation
that
today
already
exists
across
these
dispersed
set
of
tools,
it's
essentially
being
able
to
orchestrate,
right?
If
we
were
at
a
symphony,
it's
the
it's
the
conductor,
it's
orchestrating
all
of
those
actions
together
instead
of
the
rep
having
to
manually,
you
know,
switch
between
systems
to
perform
these
functions.
Now,
when
reps
do
that,
because
let's
be
real,
that's
part
of
what
they've
done
today.
That's
how
they've
been
doing
operations.
You
have
to
think
about
the
impact
of
time
and
the
impact
of
efficiency
that
tools
like
Ellie
surface
out
on
ops
for.
SPEAKER_02
11:57
So
is
it,
I
guess,
uh
kind
of
like
an
LLM,
or
is
it
maybe
like
uh,
I
don't
know,
like
an
analytics
tool,
or
is
it
maybe
a
combination
of
the
two?
SPEAKER_00
12:07
It's
a
combination.
So
as
you
think
about
the
the
DNA
of
Ellie,
right?
It
is
a
AI-driven
orchestrate.
So
there's
three
layers
to
it.
There's
an
orchestration
layer
that
happens,
which
allows
the
functions
and
the
different
tools
to
come
together
and
execute
against
those
tasks.
Today,
there's
a
lot
of
tooling
out
there
that
exists.
Well,
you'll
see,
hey,
we
could
automate
your
track
and
trace.
That's
a
specific
workflow.
Hey,
we
could
automate
your
carrier
verification,
that's
a
specific
workflow.
So
at
the
top,
as
you're
building
out
the
architecture
of
Ellie,
there
is
a
layer
that
takes
place
of
task
orchestration.
The
second
level
of
Ellie's
DNA
is
the
semantic
layer,
which
is
power
through
LMM.
Um,
and
essentially
it's
compute
compounding
value
over
time.
Every
action
that
is
that
is
occurring
on
behalf
of
the
user
within
the
tool,
Ellie
learns
from
it,
right?
So
next
time
you
have
a
load
that
you
have
to
cover
from
Chicago
to
Dallas,
as
it's
providing
outputs
back
to
the
user
through
the
outreach
workflows
that
we
have
on
which
carriers
are
the
good
ones,
not
only
the
carriers
that
have
capacity,
but
the
carriers
that
can
cover
the
load
right
there
and
then,
it
also
is
able
to
provide
the
users
with
intelligence
of
a
couple
of
things.
One,
exclusion,
which
are
the
carriers
that
maybe
have
had
bad
service
on
that
lane
in
recent
history,
which
are
uh
carriers
that
perhaps
have
not
been
approved
through
the
most
recent
verification
changes.
So
it
is
it
has
the
ability
to
learn
from
the
functions
that
a
person
has
taken,
but
also
on
its
own
is
able
to
learn
over
time.
So
that's
where
you
start
to
see
some
of
the
semantic
AI
layer
technology
and
training
of
these
of
these
models
become
very
powerful.
One
of
the
things
that
we
are
really
focusing
on
right
now
is
building
out
not
just
the
operational
memory
layer,
but
also
how
we're
incorporating
the
feedback
loops
to
retrain,
to
retrain
these
models
over
time
through
each
through
each
action
that
happens.
SPEAKER_02
14:04
Is
that
retraining
at
the
at
maybe
the
at
the
Envoy
level
or
the
LE
level?
Or
is
it
maybe
at
the
company
level
or
kind
of
a
mix
of
two?
SPEAKER_00
14:15
You
know,
it's
it's
a
mix
of
two,
right?
It's
a
particularly
specific,
you
know,
and
most
specifically
when
you're
looking
at
load
level
detail,
lane
level
detail.
So
that
is
to
to
the
training
of
functions
of
our
customers,
right?
Because
you're
gonna
have
users
where
they're
gonna
have
different
uh
guardrails
and
different
methods
of
how
they
manage
the
carriers
that
are
approved
within
their
network,
what
what
exceptions
are
deemed
to
be
exceptions,
right?
That's
also
very
common
in
operations.
You're
not
gonna
have
identical
SOPs
across
our
customer
base.
So
Ellie
has
to
ensure
that
not
only
it
learns
over
time,
but
is
able
to
execute
against
those
SOPs
that
are
unique
to
the
customer
base
that
we
are
servicing.
For
us,
where
we're
also
learning
and
what
where
Ellie,
from
a
from
a
um
technology
perspective,
is
also
ensuring
things
like
speed,
right?
It
learns
behaviors
and
patterns
of
how
we
get
faster.
If
today
it
took,
you
know,
1.8
minutes
to
conduct
this
outreach
to
get
a
carry
on
the
load,
how
what
actions
do
we
need
to
take?
Or
in
this
case,
it
observes
what
actions
a
user
may
take
to
be
able
to
continue
to
get
better
over
time.
SPEAKER_02
15:23
Within
a
brokerage,
are
there
other
use
cases
or
other
departments
that
are
are
finding
value
in
LE
or
Envoy?
SPEAKER_00
15:32
So
in
current
state,
we've
actually
been
having
a
lot
of
conversations
with
leaders
at
brokerages,
particularly
in
the
area
of
compliance.
You
know,
there's
a
lot
out
in
the
industry
that
is
solid
and
provides
really
good
solutions
in
terms
of
verifications
and
how
to
keep
up,
you
know,
with
your
hundreds
and
hundreds
of
rows
of
databases,
of
carriers,
documentations,
how
to
go
about
onboarding,
et
cetera.
But
one
of
the
things
that
we're
trying
to
do
is
change
the
or
bring
the
conversation
to
the
table
of
how
do
you
ensure
that
all
this
work
and
all
this
investment
that
you're
doing
through
these
platforms
gets
executed
against,
right?
So
that
is
one
interesting
piece
for
us
that
we're
seeing
in
the
evolution
of
ruling
this
out,
where
originally
you
could
say
the
thesis
was
on
the
ops
floor,
the
impact
that
this
has
with
operational
efficiency
and
productivity.
We're
taking
it,
we're
continuing
to
evolve
that
conversation
and
having
those
dialogues
with
the
compliance
teams.
The
compliance
teams
are
now
looking
at
it
and
saying,
wait
a
minute,
this
is
a
really
good
way
of
being
able
to
one,
drive
continuity
and
how
operations
is
getting
executed,
but
two,
it's
able
to
provide
a
higher
level
of
confidence
that
if
they
have
carriers,
let's
say,
marked
in
a
uh
do-not-use
bucket,
right,
that
they're
not
going
to
be
going,
you
know,
pursued
after
just
for
the
sake
of
covering
a
load.
So
I
find
that
very
interesting.
SPEAKER_02
16:55
And
so
from
I
I
guess
um
an
onboarding
perspective,
you
know,
for
a
lot
of
companies,
I
would
imagine
that
they're
dealing
with,
you
know,
messy
data
or
unorganized.
You
know,
they
they
talk
about
how
you
know
dirty
data
is
expensive.
And
I
I'm
curious
as
to
what
a
brokerage
could
do
to
maybe
start
cleaning
up
some
of
that
data
so
that
they
can
get
themselves
prepared
to
start
taking
advantage
of
tools
like
Envoy
and
Ellie.
SPEAKER_00
17:22
Yeah.
So
some
of
the
things
as
as
you
know,
prospects
or
our
brokerages
are
evaluating
implementation
of
technologies
such
as
ours,
some
of
the
areas
to
look
at
for
sure,
you
know,
evaluating
how
clean
and
how
current
the
data
set
is.
We
all
know
that
what
happened,
you
know,
in
in
our
industry
30,
30
days
ago
may
already
be
stalled.
So
what
are
a
lot
of
this
has
to
do
with
using
these
types
of
technologies
also
as
an
opportunity
to
establish,
renew,
or
create
operating
procedures
at
these
companies.
So
one
of
the
first
areas
is
what
is
your
what
is
your
audit
timelines
with
your
carriers
in
regards
to
compliance?
Number
two,
what
are
the
different
thresholds
and
customer
level
information
that
you
have
built
into
your
TMS?
Some
of
these
things
could
be
special,
uh,
you
know,
special
instructions,
special
requirements
that
key
customers
may
have
that
impact
the
way
that
a
load
is
getting
booked.
And
we
start
to
see
this,
you
know,
master
data
is
such
an
important
part
of
my
career,
also
in
transportation
management,
but
it's
it's
also
very
relevant
where
I'm
sitting
today
because
if
you
think
about
it,
the
knowledge
of
I'm
gonna
ship
to
the,
you
know,
I'm
shipping
this
from
my
key
customer,
they're
gonna
need
X,
Y,
and
Z.
That
lives
in
people's
heads.
To
be
able
to
train
models
like
these
to
understand
and
to
be
able
to
perform
in
the
same
function
that
a
human
would,
you
have
to
make
sure
that
the
master
data
and
core
systems
like
TMSs
are
up
to
date.
So
we
do
a
lot
of
conversation
ahead
of
a
launch
or
you
know,
even
before
a
full
implementation
happens,
where
we
sit
and
go
through
a
framework
of
master
data
across
compliance,
operations,
track
and
trace.
And
then
we
also
ensure
that
working
with
the
operators,
if
there's
different
guardrails
or
SOPs
that
are
need
to
be
implemented,
that
we
also
are
aware
of
what
those
are.
SPEAKER_02
19:21
Yeah,
because
I
I
think
I
heard
uh
you
know
recently,
and
I
correct
me
if
I'm
wrong,
but
you
know,
in
order
to
make
sure
that
you
have
proper
clean
data,
not
just
you
know,
from
all
uh
gathering
it
in
and
initially
setting
that
up,
but
also
making
sure
that
almost
like
if
this,
not
a
if
this
then
that,
but
this
equals,
you
know,
this
column
title
is
the
same
thing
as
this
column
title
and
this
spreadsheet,
and
just
making
sure
that
those
two
things
match
and
what
uh
the
carrier
is
is
quantifying
in
a
certain
column
is
the
same
thing
that
the
finance
department
wants
to
quantify
in
another,
you
know,
spreadsheet
report
that
they
might
be
looking
at.
But
the
bare
minimum
is
that
those
columns
and
that
that
data
is
labeled
the
same
thing.
Am
I
understanding
that
correctly?
SPEAKER_00
20:07
So
the
the
yeah,
the
data
normalization
is
is,
I
think,
a
key
to
success,
right?
You
need
to
make
sure
that
you,
you
know,
brokerage
leaders
or
any
leaders
that
are
evaluating
how
to
implement
and
adopt
these
types
of
technologies,
that
they
also
understand
that,
yes,
they're
very
easy
to
deploy.
That,
you
know,
the
the
timeline
really
varies
on
when
they
want
this
live,
but
to
do
it
right
does
take
some
pre-work,
right?
That's
where
we
want
to
make
sure
our
customers
understand
this
is
not
a
rollout
for
the
sake
of
rollout.
We
want
to
make
sure
that
as
part
of
these
deployments,
the
conversations
are
happening
with
finance,
with
compliance,
with
your
leaders
across
different
uh
shared
service
groups
like
track
and
trace.
There's
also
a
big
piece
of
this
that
not
a
lot
of
people
are
talking
about,
which
is
the
change
management
side
of
it,
right?
Getting
the
technology
right,
getting
the
maps
right,
getting
the
data
normalized,
all
of
that
is
important.
It
all
will
fail
if
there's
no
change
management
strategies
that
are
discussed
at
the
table
and
not
just
with
your
key
stakeholder
that
is
looking
to
bring
this
in,
but
cross-functionally
at
corporations
to
ensure
that
everybody
buys
in
not
only
into
the
outcome
and
the
success
that
this
is
going
to
bring
from
a
tactical
operational
side,
but
also
what
it's
gonna
do
for
them
as
a
company.
SPEAKER_02
21:26
Yeah,
that
that
makes
a
ton
of
sense.
It's
just
getting
everybody
on
board
and
speaking
the
same
language
is
sometimes
the
hardest
part
of
the
entire
process,
but
you
can't
get
to
that
end
goal
and
you
can't
get
to
those
actionable
insights
unless
everybody
is
is
rowing
together
in
in
the
same
direction.
Uh
I'm
curious
from
the
I
guess
the
psychological
level
of
it
where
you
know
there's
a
lot
of
fear
around
AI.
And
that
that's
one
thing
that
I'm
discovering
in
a
lot
of
these
conversations
is
like,
yes,
that
fear
exists,
but
there's
also
an
incredible
amount
of
demand
for
people
that
need
help
with
this
massive
amount
of
data,
these
massive,
you
know,
tech
not
or
I
guess
tech
budgets
that
are,
you
know,
purchasing
all
of
these
different
tools.
How
to
how
to
figure
out
where
to
spend
your
time
and
energy?
And
so
creating
that
level
of
clarity.
So
when
you're
you're
first
implementing
and
and
or
first
onboarding
a
client,
how
important
is
it
to
have
you
know
those
carrier
sales
reps,
the
the
in-the-trenches
employees
who
are
gonna
buy
into
it
and
not
push
back
against
some
of
this
change
management
that's
required
of
them?
SPEAKER_00
22:32
Yeah.
Um,
very,
very
interesting
question
um
because
it's
very
much
real.
The
pushback
in
my
experience,
I
can't
speak
up
for
everybody,
but
I
often
find
that
the
pushback
isn't
about
the
technology
itself.
The
pushback
is
about
the
trust
and
the
uncertainty
of
what
it's
going
to
do
for
the
roles
that
essentially
are
getting
asked
to
use
this
technology.
And
some
of
the
some
of
the
experiences
just
in
in
the
time
that
I've
been
with
some
of
our
customers
is
really
figuring
out
is
this
gonna
create
more
work,
right?
Is
it
can
I
trust
it?
Is
it
gonna
know
what
to
do
without
me
having
to
tell
it
to
what
to
do?
Uh,
what
happens
if
there's
a
mistake?
Who's
gonna
be
held
accountable?
So
there's
a
lot
of
very
valid
questions
that
take
place
when
you
start
the
conversations,
which
I
love
and
I
find
fascinating
because
what
I
love
to
see,
and
I
think
you
know,
you
look
at
our
team,
and
what
we're
most
proud
of
is
When
these
questions
surface
at
the
start
of
a
project,
let's
say
fast
forward
two
weeks,
three
weeks,
you
know,
as
an
example,
and
you
see
the
smiles
on
their
face,
you
see
the
aha
moment
of,
hey,
you
know,
you
told
me,
but
I
really
wasn't
sure.
As
a
product
leader,
that's
probably
one
of
the
most
exciting
parts
of
it.
The
success
of
this
is
to
be
very
intentional
in
how
you're
rolling
it
out.
So
the
conversations
that
we
have
with
our
executive
sponsors
and
our
leaders
that
are
evaluating
the
investment,
it's
also
ensuring
that
their
team
is
set
up
for
success
by
being
intentional
in
the
rollout.
Anybody
that's
coming
with
these
types
of
projects
or
tools
that
say,
hey,
you
know,
we
could,
we're
turning
it
on
and
we're
turning
it
off
for
everything,
that's
just
a
miserable,
you
know,
way
to
allow
the
people
that
are
impacted
by
these
changes
to
fail
because
it's
overwhelming,
it's
it's
new,
it's
not
calibrated.
So
for
us,
the
way
we
approach
it
is
ensure
that
we're
intentional
about
what's
the
workflow
that
we're
gonna
be
addressing,
then
what
happens
next,
and
then
mutually,
how
do
we
define
success?
And
I
think
where
we
have
seen
a
big
shift
in
the
willingness
to
adopt
these
tools
is
when
we're
able
to
sit
down
with
the
folks
that
are
gonna
be
held
accountable
to
use
it
and
also
ask,
what
do
they
care
about?
What's
success
to
them
look
like,
and
being
able
to
prove
with
the
real
results
that
they're
getting
those
outcomes.
SPEAKER_02
24:57
Oh,
that
that's
super
interesting.
And
so
as
you
know,
the
the
employees
are
starting
to
use
the
platform
and
they're
they're
finding
success
with
it,
what
happens
from
I
I
guess
the
the
training
part
of
it,
the
feedback
part
of
it,
because
there
is
you
know
a
lot
of
concern
around,
you
know,
well,
what
if
it
hallucinates
and
what
if
it
you
know
it
gets
something
wrong?
And
you
know,
uh
what
does
the
process
look
like
then?
Is
it
their
team
letting
you
know
and
then
you
you
you
fix
it
in
the
system,
or
is
it
maybe
like
a
root
cause
issue
where
there's
you
know
something
is
wrong
with
their
data?
What
does
that
process
look
like?
SPEAKER_00
25:32
Yeah,
so
our
our
particular
I'll
tell
I'll
talk
to
you
a
little
bit
about
our
specific
process
because
everybody
may
be
approaching
this
um
differently.
The
first
thing
is
during
our
implement
our
our
implementation
framework,
we
spend
a
lot
of
time
preparing
for
production
deployment.
So
there's
already
a
level
of
security
and
assurance
of
what
the
product
experience
is
going
to
look
like.
So
we
do
go
through
various
user
testing
stages
um
but
prior
prior
to
full
rollout.
Once
it's
rolled
out,
we
also
ensure
that
there
are
resourcing
that
is
allocated
to
monitor
not
only
the
activity
that
is
happening,
but
also
that
we
have
well-defined
plans
if
for
whatever
reason
we
need
to
we
we
need
to
shift
our
approach.
And
that
shift
would
be
turning
it
off
in
production.
We
recognize
the
severity
of
impacting
freight
in
production.
So
we've
established
protocols
where
we
can,
if
we
need
to
up
to
this
point,
knock
on
wood.
We
would
not
have
to
do
that,
but
if
we
needed
to
pull
back,
we
can.
Now,
what
is
a
day
in
a
life
and
experience
look
like?
Think
of
week
one
of
your
goal
life.
You're
gonna
have
somebody
that
is
on
the
floor
with
your
team
members,
not
only
being
there
for
training
or
also
for
any
type
of
potential
troubleshooting
that
may
come
on.
We
have
a
support
team
that
is
also,
you
know,
on
standby,
waiting
to
monitor
any
defects
or
any
issues
that
may
surface.
But
strategically,
the
way
that
we
approach
this
is
we're
doing
customer
engagement
sessions
throughout
the
hyper
care
period
of
the
goal
life,
where
we're
working
with
our
stakeholders,
not
only
providing
them
with
metrics
of
what
we're
seeing,
how
the
product's
behaving,
um,
you
know,
some
of
the
outcomes
that
we
go
in
with
is,
you
know,
how
many
loads
have
been
covered,
how
many,
how
many
offers
have
been
accepted,
how
many,
for
example,
the
the
the
time
to
cover,
some
of
those
metrics
where
they're
seeing
week
one,
week
two
metrics.
So
it
helps
them
get
confidence.
Now,
you
mentioned
something
about
is
it,
is
it,
you
know,
are
the
users
able
to
provide
your
feedback?
Absolutely.
For
us,
we
believe
that
the
people
that
use
it
are
the
ones
that
are
going
to
give
you
the
best
source
of
feedback.
So
we
have
established
channels
where
they
could
email
us
and
say,
hey,
you
know,
today
I
was
working
on
this
and
I
wish
Ellie
could
do
this.
Or
hey,
Michelle,
you
know,
I
was
working
on
this
and
I
wasn't
really
sure
about
this
button.
Of
course,
hey,
Michelle,
it
doesn't
come
to
me,
it
goes
to
our
support
group.
But
things
like
that
where
you'd
be
surprised
through
those
feedback
channels,
we
get
some
really
exciting
ideas
that
we
can
even
incorporate
into
a
broader
roadmap.
SPEAKER_02
28:04
That
yeah,
I
think
that
that's
where,
personally
speaking,
that
that's
where
I
find
the
biggest
issue
with,
you
know,
uh
ever
on
the
large
language
models
is
that
it
allows
me
to
do
so
much.
So
now
focus
becomes
the
the
primary
objective
instead
of
being
pulled
in
a
million
different
directions
because
I
can
do
so
much
more.
And
so
I
I
I'm
curious
from
the
I
guess
the
the
setup
of
an
ideal
brokerage.
Is
there
one
person
in
charge
of
sort
of
AI
operations,
or
is
it
maybe
a
new
department
that
that's
forming
within
the
organization,
or
is
it
the
traditional
employee,
traditional
departments,
but
they
have
that
sort
of
AI
superpower
added
onto
them?
SPEAKER_00
28:45
I
wouldn't
say
that
it
has
to
be
somebody
that
is
focused.
We
are
seeing
where
there
are
uh
companies
that
are
designating
you
know
special
projects,
uh
liaisons
to
take
on
projects
like
these,
which
I
think
is
great
because
they're
focused.
They're,
you
know,
it's
that
part,
they
don't
have
to
choose
between
rolling
out
AI
and
covering
freight,
right?
Because
that's
also
could
be
risky
if
they're
not
um
focused
on
the
deployment
and
the
implementation
and
deployment
of
the
of
the
technology.
But
in
the
case
that
they
don't
have
the
benefit
of
these
resources,
that's
where
we
come
in,
right?
We're
able
to
be
their
um
their
point
of
contact.
We're
able
to
help
them
with
strategy,
not
only
in
change
management,
but
also
how
to
how
to
implement
it.
And
you
know,
you
have
to
make
sure
that
we
also
establish
clear
protocols
of,
and
you've
heard
this
before,
our
strategies
of
human
in
the
loop.
At
the
end
of
the
day,
the
end
goal
is
to
be
able
to
drive
type
plus
automation,
right?
Be
able
to
trust
that
that
the
tool
is
performing
where
it
needs
to
be
to
be
fully
autonomous.
It
takes
time
to
get
there,
right?
And
up
to
this
point,
we
need
to
make
sure
that
there
is
buy-in
on
how
for
the
language
for
the
language
model,
what
feedback
we
need
from
the
humans,
right?
To
be
able
to
train.
And
we
talked
about
semantically,
or
we
talked
about
how
it
observes
those
phases
in
these
types
of
projects
become
very
important
because
then
you
also
have
to
make
sure
you
define
it
when
the
human
in
the
loop
ends
and
it
falls
to
full
uh
autonomy.
SPEAKER_02
30:14
Yeah,
I
I
think
that
that's
a
really
great
point
because
it's
it's
one
thing
to
adopt
these
technologies,
but
it's
it's
a
completely
different
other
to
not,
I
guess,
ignore
it
or
use
all
of
the
features.
Because
I
think,
you
know,
for
a
modern
uh
SAS
product,
it's
something
crazy
where
only
like
20%
of
the
features
are
actually
utilized,
and
that's
a
good
case
scenario.
Um
I
think
you
know,
getting
them
on
boarded,
getting
them
those
early
light
bulb
moments,
I
think
that
that
that's
really
key.
And
I
for
a
lot
of
companies,
it
might
be
really
hard
for
them
to
document,
initially
document
their
processes
where
they
they
may
have
never
done
this
before.
And,
you
know,
thinking
about
it
in
a
different
lens
because
of
this,
you
know,
XYZ
reasons
why.
And
so
I
think
that
that
that's
a
challenge
too
for
a
lot
of
companies.
And
I'm
I'm
curious
if
if
that
is
a
prevalent
issue,
or
maybe
it's
you
know,
kind
of
case
by
case.
SPEAKER_00
31:05
No,
I
think
it's
you
know,
it's
a
it's
a
case
by
case.
It
also
you
but
it's
all
gonna
be
dependent
on
the
mature
and
the
operational
maturity
that
some
of
these
companies
have,
right?
And
the
the
key
is
to
be
able
to
be
flex
as
a
as
a
um
technology
provider,
to
be
able
to
be
flexible
in
any
of
those
environments,
right?
Being
able
to
have
the
right
strategy
and
frameworks
to
support
companies
that
have
their
data
up
to
date,
have
all
the
SOPs
documented,
understand
each
and
every
single
one
of
their
customer
profiles.
But
just
as
important
to
walk
in
the
door
with
somebody
that
says,
I
haven't
looked
at
this
and
you
know,
I
need
a
partner.
Who
do
you
guys
partner
with?
Who
can
help
me
improve
my
compliance?
Who
can
help
me,
you
know,
create
a
standard
for
track
and
trades,
like
things
like
that,
where
the
fact
that,
you
know,
when
you
look
at
the
team
at
Envoy,
and
one
of
the
reasons
that
also
drove
me
to
move
in
is
these
are
folks
that
have
done
the
job.
I'm
I'm
sitting
peers
of
mine,
uh,
partners
and
team
members
that
I
can
have
a
conversation
as
we're
thinking
product
strategy
that
bring
the
lens
of
what
it
was
like
many
years
ago
for
them
to
do
the
job.
And
I
think
that
also
changes.
And
when
we
have
those
conversations
where
we
present
those
frameworks,
it's
not
just
tech-based,
right?
It's
really
getting
into
the
weeds
almost
like
we're
speaking
their
language
because
many
of
those
have
done
the
job.
And
that
also
adds
to
a
greater
level
of
credibility
of
saying,
okay,
these
guys
are
not,
you
know,
understand
what
it's
like
to
jeopardize
and
miss
pickup,
right?
They're
not
gonna
come
in
here
and
mess
me
up
because
their
tool
is
not
gonna
work.
So
that's
just
a
little
extra
to
add
because
I
do
think
that
it
that
it
changes
the
conversation.
And
at
the
end
of
the
day,
for
us
and
and
as
we
look
at
our
product,
everything
that
we're
doing,
we
understand
one,
relationships
matter.
Two,
you
can't
mess
up
freight.
You
have
to,
one
bad
load
could
could
could
jeopardize
the
relationship
with
the
shipper.
And
we
don't
want
our
brokerage
partners
to
go
through
that,
right?
So
that's
why
we're
very
intentional,
very
cautious
in
how
we
roll
it
out.
If
a
customer
comes
in
and
you
know
wants
to
have
a
very
aggressive
timeline
of
how
to
deploy
this,
we're
gonna
be
the
ones
that
are
gonna
say,
hey,
listen,
have
you
thought
about
this
other
angle
for
the
sake
of
your
operation?
And
that's
where
we
move
from
just
being
another
provider
to
becoming
a
partner
in
their
journey
of
AI.
SPEAKER_02
33:19
Yeah,
I
I
think
that
that
that's
a
a
really
brilliant
way
to
I
got
a
couple
more
questions
for
you,
but
I
think
that
that's
a
perfect
way
to
kind
of,
you
know,
come
to
the
the
finish
line
of
the
of
this
conversation
because
there's
so
many
tech
companies
that
are
seeing
transportation
and
logistics,
and
they
think
that
it's
such
a,
oh,
well,
we
can
just
go
in
there
and
disrupt
it
and
you
know,
uh
become
a
unicorn
business.
And
it's
it's
you
gotta
have
people
who
speak
the
language
who
know
how
critical
different
aspects
of
the
transportation
process
is.
And
it
sounds
like
you
know
you
you
guys
have
really
built
you
know
a
strong
team
over
there.
And
I'm
curious,
um,
I
mentioned
last
couple
questions.
Is
it
are
you
focused
mainly
on
enterprise
you
know,
brokerages
right
now,
or
maybe
mid-size?
Is
there
room
for
maybe
you
know
some
SMBs
to
kind
of
take
advantage
of
this?
Or
is
that
maybe
in
the
product
roadmap?
SPEAKER_00
34:08
So
right
now,
you
know,
mid-size
enterprise
is
is
really,
you
know,
we
have
we
have
a
brilliant
team,
a
commercial
team
that
is
looking
at
all
the
different
angles,
right,
to
be
able
to
tackle.
And
that's
as
a
product
leader,
the
cool
part
of
my
job,
right?
That
my
product
doesn't
have
to
be
isolated
to
just
one
user
journey
and
being
able
to
plug
and
play
where
our
customers
need.
That's
that's
the
exciting
part.
SPEAKER_02
34:33
All
right,
Michelle.
What
is
one
thing
that
you
think
is
important
to
mention
that
we
haven't
already
talked
about?
The
one
thing,
man,
that's
a
hard
one.
SPEAKER_00
34:41
Um,
for
your
audience
and
your
viewers,
this
is
not
about
humans
versus
AI.
And
I
know
that
sounds
a
little
nerdy
of
me
to
say,
but
I
mean
it
from
the
heart.
The
future
is
not
humans
versus
AI.
It's
really
around
what
operational
teams
can
do
with
AI
and
how
AI
can
become
a
competitive
advantage
in
the
market
when
they're
servicing,
whether
you're
servicing
a
shipper,
whether
you're
a
BPO
that's
servicing
a
logistics
provider,
um,
or
you're
just
a
broker
too
in
your
own,
right?
It's
about
how
you
could
use
these
technologies
to
help
you
outperform
your
other
call,
your
other
buddy.
SPEAKER_02
35:24
No,
I
I
love
it.
I
I
am
uh
a
tech
optimist,
and
so
I
I
agree.
And
that
was
one
of
the
things
with
coming
into
this
series.
I
just
wanted
to
uncover
is
is
are
people
actually
scared,
or
is
that
what
I'm
hearing
in
the
news?
And
I
think
it's
the
former
rather
than
the
latter.
So
really,
really
appreciate
your
your
time
and
your
your
expertise
and
sharing
it
with
the
audience
today.
Where
can
folks
connect
with
you?
Uh,
book
a
demo
with
with
Envoy
and
Ellie
and
and
all
that
good
stuff.
SPEAKER_00
35:51
Yes,
come
meet
Ellie.
We
want
to
hear
from
you
guys.
Our
website,
um,
tryonvoy.ai.
And
then
um,
of
course,
reach
me
out,
reach
out
out
to
me
on
LinkedIn
if
you
have
questions,
if
you
just
want
to
challenge
some
of
the
points
that
that
they
heard
about
today,
I'm
always
up
for
a
healthy
debate.
And
uh,
we
have
a
great
team
of
experts
as
well,
right?
I
uh
I
don't
tend
to
be
the
AI
know
it
all.
Um,
16
years
does
give
me
some
credibility.
So
I'm
excited
to
have
the
conversation
and
help
address
some
questions
on
change
management.
It's
not
just
about
our
specific
tech,
but
if
there
are
people
that
are
starting
to
think,
you
know,
whether
you're
in
the
early
journey
of
sourcing
a
technology
product
or
you're
in
the
middle
of
figuring
out
how
to
best
implement
it,
we're
an
open
book
and
we're
happy
to
help.
SPEAKER_02
36:38
That
is
a
perfect
place
to
end
it.
So,
Michelle,
thank
you
so
much.
Thank
you.
Appreciate
it.
SPEAKER_01
36:48
Thanks
for
tuning
in
to
another
episode
of
Everything
Is
Logistics
where
we
talk
all
things
supply
chain
for
the
thinkers
in
freight.
If
you
like
this
episode,
there's
plenty
more
where
that
came
from.
Be
sure
to
follow
or
subscribe
on
your
favorite
podcast
app
so
you
never
miss
a
conversation.
The
show
is
also
available
in
video
format
over
on
YouTube
just
by
searching
Everything
is
Logistics.
And
if
you're
working
in
freight
logistics
or
supply
chain
marketing,
check
out
my
company
Digital
Dispatch.
We
help
you
build
smarter
websites
and
marketing
systems
that
actually
drive
results,
not
just
vanity
metrics.
Additionally,
if
you're
trying
to
find
the
right
freight
tech
tools
or
partners
without
getting
buried
in
buzzwords,
head
on
over
to
CargoRex.io
where
we're
building
the
largest
database
of
logistics
services
and
solutions.
All
the
links
you
need
are
in
the
show
notes.
I'll
catch
you
in
the
next
episode
and
go
Jags.
