►
Description
Kava Labs Lead Engineer Kevin Davis (https://twitter.com/@KevinJYDavis) and NEAR Co-Founder, Illia Polosukhin (https://twitter.com/ilblackdragon) do a deep dive on cross-chain DeFi, stablecoins, and how Kava’s technology works in the 31st episode of the Whiteboard Session. Kava Labs is a developer of the Kava blockchain, the first cross-blockchain DeFi platform.
------- About Whiteboard Series ---------
The Whiteboard Series with NEAR is a set of video interviews with Founders who are building scalable solutions for Blockchain. We deep dive into their story, what they're developing and how it is helping the ecosystem.
Follow the latest from NEAR Protocol on,
Discord: https://near.ai/discord
Medium: https://near.ai/medium
Twitter: https://near.ai/twitter
GitHub: https://near.ai/github
#WhiteboardSeries #Blockchain #FutureIsNEAR
B
B
So
that's
sort
of
how
I
got
started
and
let's
take
it
from
there
all
right
all
right,
so
I
think
the
the
place
to
start
is
sort
of
what
is
the
fundamental
unit
of
defy
right
now,
and
that
is
the
over
collateralized
loan
and
so
I'm
gonna
go
through
the
story
of
an
over
collateralized
loan
and
then
we'll
see
if
we
can
sort
of
get
to
all
the
technical
details
and
dive
into
where
these
things
break
and
how
they
work
sounds
good.
All
right.
So
fundamentally,.
B
Someone
controls
an
address
here
with
some
amount,
X
coins,
and
it
would
be
nice
right
if
we
could
put
our
coins
to
work
and
sort
of
a
lot
of
people.
Think
of
various
different
ways
that
we
could
do
that
and
one
of
the
ones
that's
very
hot
right
now
is
defy
door
over
collateralized
loans,
and
what
I'd
like
to
think
of
this
is
just
an
automated
loan
facility
that
sort
of
can
be
run
relatively
autonomously
and
we'll
get
into
where
it's
not
autonomous.
B
B
And
and
some
protocols
you
need
to
like,
convert
your
coins
into
a
pool
of
coins
and
other
protocols,
you
don't
the
one
we
work
on
you
don't
so
I
can
imagine.
I
can
just
transfer
my
coins
X
and
we'll
call
it.
My
coins
in
CBP
and
I'm,
just
gonna
use
the
CDP
term,
apparently
we're
doing
like
rebranding
and
we're
calling
them
like
safes
now
or
something
but
I.
B
So
I
didn't
I,
didn't
change
the
ownership,
but
this
box
here
is
now
capable
of
creating
a
bunch
of
financial
contracts
that
can
be
run
on
a
blockchain,
and
so
the
first
sort
of
thing
we
want
to
be
able
to
do
with
this
is
to
take
my
balance
X
and
to
create
a
coin
off
of
X,
and
so
right,
I
can't
probably
create
more
x,
so
would
be
surprising
right
protocol.
Just
let
me
create
more
ex
probably
x
would
have
an
inflationary
problem,
so
instead
we're
gonna.
B
B
B
First,
we
have
to
figure
out
how
much
debt
can
I
draw
off
of
X
and
then
why
anyone
would
do
this.
So
first
question
has
to
be
I
need
to
know
the
value
of
x
acts
in
dollars
to
USD
and
I
need
to
be
able
to
get
there
with
high
frequency
like
it
can't
be
like
that.
If
this
value
is
changing
a
lot,
I
need
it
to
change
a
lot
here
and
then,
because
typically
X
is
a
very
illiquid
super
highly
volatile
asset.
B
B
B
Wants
to
write
it
down,
nobody
wants
to
drive
it
up
because
it's
already
super
economically
yeah.
It's
not
a
common
set
up
and
financial
things.
I
think
one
thing
that's
worth
pointing
out
is
like
the
traditional
thing
you
do
with
collateral.
As
you
post,
highly
liquid
collateral
to
draw
an
illiquid
debt
or
sorry,
you
take
illiquid
assets
and
draw
a
highly
like
yeah
debt.
Here
we
have
two
illiquid
assets
and
so
we're
we're
not
in
a
traditional
set
up
and.
B
Or
you
have
a
very
small
number:
lateral
X's,
ten
yeah,
yeah
and
I
can
use
X
to
make
many
different
loans
right.
Alright,
so
we
we've
we've
agreed
that
I
can
draw
some
amount
of
X
here
and
we're
gonna
allow
this
system
to
be
replicated
both
across
the
asset,
X
and
then
across
any
other
assets.
So
it's
going
to
go
deep
and
wide
in
terms
of
many
assets
that.
B
B
C
B
Total
sum
of
all
debt
will
be
these
fungible
coins
and
the
total
sum
of
all
this
is
the
collateral
backing
those
fungible
coins
and
that's
what
gives
us
the
expectation
of
dollar,
parity
and
I,
say
expectation,
because
I
cannot
take
dollar
coin
dollar
debt
coin
turn
it
back
into
collateral.
There's
no
in
protocol
thing
that
just
swaps
those
around
and
there
is
no
guarantee
that
this
thing
will
ever
settle.
So
my
maker
has
sort
of
what
they
call
emergency.
You.
C
D
B
Is
the
person
that's
doing
this?
It's
probably
someone
who
has
way
too
much
if
they
literally
have
a
problem
of
I
have
too
much
II,
and
this
is
something
I
can
do
with
it
yeah,
and
so
what
I
sort
of
like
to
look
at
this
system
here
is
this
thing
that
is
controlled
by
the
the
debt
ceiling?
It's
gonna
be
pretty
hard
to
scale
right
with
a
single
asset
mm-hmm
and
finding
good
XY
and
Z
candidates
is
very
difficult.
B
A
B
C
A
B
D
B
Okay,
so
users
interact
with
the
system.
They
create
these
collateralized
loans,
loans,
accumulate
fees
which
are
controlled.
The
fees
themselves
be
controlled
by
governance
when
users
close
positions,
they
pay
those
fees.
In
the
governance
token,
so
you
have
both
inflationary
and
deflationary,
and
then
we
have
lender
of
last
resort.
B
C
B
C
B
B
C
B
Amortized
trust
like
IBC
you're,
still
going
through
places
that
require
you
to
trust
the
Federation
itself,
the
sort
of
relayer
there
has
to
be
one
honest
trailer,
but
we're
much
less
than
sort
of
there's
three
signers
on
my
point,
and
so
we
think
this
is
a
really
cool
construct
right
here
that
if
I
can
take
an
application-specific
blockchain
and
could
be
a
Dex
could
be
a
sort
of
I
would
be
skeptical.
There
are
money
tokens
in
application-specific
chains,
but
Dex
is
things
that
are
useful
and
tradable.
It
could
be
on
application-specific
block
chains.
C
A
B
A
C
B
A
B
Think
about
CD
piece
today,
it's
like
you
open
a
dashboard
and
it
looks
like
a
finance
dashboard
and
you
open
a
position
and
then
you
way
over
collateralize
it
because
you're
worried
about
the
fees
and
how
long
it's
gonna
take,
and
you
don't
want
to
worry
about
it.
If
it's
just
a
button,
that's
like
lever
up
whatever
you're
doing
right
here.
In
the
background
like
CDP
gets
opened,
a
stable
deck
gets
drawn,
an
automatic
conversion
happens
through
something
like
Yunus,
whap
and
boom.
You
just
have
more
of
X
mm-hmm.
A
B
A
B
B
Y
can
be
seized
by
some
outside
authority.
Then
how
do
I
sort
of
judge
this
debt
here
yeah?
If
I
really
don't
want
to
be
exposed
to
Y,
or
at
least
I,
don't
want
to
I.
Don't
want
my
users
to
have
to
think
about
sort
of
exposure
to
to
why
mm-hmm,
if
Y
isn't
pretty
trusted
and
then
I
would
argue-
or
at
least
our
philosophy
is
like,
you
should
really
avoid
these
well.
A
So
my
question
is
in
the
world
of,
like,
let's
say,
IBC
or
any
kind
of
like
cross-chain
stuff,
as
a
user
like
I
need
to
at
least
have
some
understanding
where,
as
the
origin
of
the
like,
if
it's
just
like
I'm,
using
all
the
examples
like
a
cryptic
ad
right
on
my
chain.
But
it
actually
was
from
another
chain
right.
Yep
like
using
it
here.
Kind
of
requires
me
to
understand
a
little
bit
about,
especially.
A
B
B
This
is
what
we
want
to
do
here,
and
so
our
approach
to
prices
is
the
same
as
staking,
which
is
that
the
people
who
govern
the
token
should
be
subject
to
slashing
conditions
both
for
the
security
of
the
chain
and
for
the
prices
that
are
fed
into
the
chain.
All
right,
and
so
the
the
conditions
for
slashing
and
atom
is
like
double
signing.
Mm-Hm.
C
B
B
C
C
B
A
A
A
B
A
A
B
B
B
A
A
A
B
B
C
A
B
B
D
C
B
B
B
C
B
Have
a
ten
thousand
dollars
with
the
program
we'll
call
that
a
lot
and
then
the
idea
is
because
I
don't
have
to
spend
transaction
fees
biting
I.
Don't
really
need
to
pay
my
keepers
here,
yeah
I,
don't
need
to
incentivize
here.
So
what
actually
happens
is
you
kick
an
auction
off
and
this
will
almost
certainly
present
arbitrage
opportunities.
C
C
A
A
B
C
B
Going
to
bid-
and
then
if
that
does
not
cross
out
the
sixty-six
debt,
then
we
have
a
really
bad
auction
and
that
accumulates
to
the
inflationary
table.
So
now
I
have
two.
So
let's
imagine
that
this
auction
raises
fifty
mm-hmm
us
DX
yeah
I
now
have
16
that
is
on
us
DX
that
is
uncovered
and
now
I'm
an
auction
table
for
that.
So
here
I'm
actually
in
collateral
mm-hmm
here
I'm
auctioning
cover
so.
A
A
A
A
C
B
A
B
A
B
A
A
A
So
how,
as
a
like,
let's
say,
I'm
I
am
market
participant.
I
do
need
to
evaluate
like
both
which
portfolio
to
hold
to
actually
do
this,
as
well
as,
if
I'm
holding
it
use.
Dx
like
when
I
want
to
close
a
position,
oh
yeah.
So
what
one
thing
we
didn't
talk
about
is
when
I
want
to
pay
it
back
right,
yep,
so
I
paid
back
I
can
only
close
my
own
CDP
or
correct.
We
did
not
have
like.
B
A
B
C
D
A
A
A
Right
yeah
so
going
back
to
how
to
evaluate
the
risk
pretty
much
on
the
whole
when
you
have
like
a
mixed
bag
of
different
assets,
a
grab
bag,
a
grab
bag
of
different
assets,
yeah
with
different
risk
profiles.
Different
like
you
may
get
like
you
know,
I
wish
me
it
stall
stall
because
it
like
something
happen
and
that
chain
can
yeah.
B
B
C
B
B
B
B
Yeah
over
time,
you
don't
want
to
I
would
say
what
you
want
to
start
with.
Is
a
smaller
debt
limit,
then
sort
of
the
risk
tolerance
of
all
of
these
assets?
You
don't
want
to
be
fully
exposed,
but
yes,
over
time,
I
think
the
actual
value
or
the
actual
alpha
during
it
generated
by
any
systems
having
governors
who
know
what
they're
doing
mm-hmm
having
governors
who
are
pricing
risk
properly.
A
A
A
B
B
C
B
A
A
B
A
B
B
A
B
A
We
were
thinking
about
it
because,
like
that's
he's
just
like
in
maker
I
do
have
I
mean
it's
the
same
idea
like
you
have
maker.
You
have
ether
like,
but
you're
joining
them
into
one.
It
makes
sense,
but
then
you
also
have
this
like
a
ton
of
the
liquid
asset.
That
is
city
there,
which
does
have
security
like
it
does
have
a
like
sound
risk
on
it,
yep
being
staked,
but
at
the
same
time,
at
least
from
practice.
We've
seen
what
this
risk.
A
A
A
B
B
C
B
Hard
to
differentiate
yourself,
it's
easy
to
set
up
a
system
that
doesn't
double
sign.
Liveness,
isn't
really
that
important
as
well.
It's
really
easy
to
be
live
with
servers
on
the
internet.
Governance
is
an
area
where
you
can
see
differentiation
among
orgs
and
I.
Do
think
that
that
is
sort
of
an
actual,
interesting
property
of
this
system
that
orgs
find
themselves
unable
to
differentiate
in
this
category
can
find
themselves
able
to
differentiate
in
all
of
these
areas
of
because.