►
Description
Lea, Lucas, and Martin from Centrifuge answer questions from the community about tokenizing real-world assets (RWA) and how the team plans to make RWA more accessible to traditional business and finance.
First AMA with Centrifuge/Tinlake - https://youtu.be/5mnghrGB4Yc
Centrifuge/Tinlake:
Website - https://centrifuge.io/
Website - https://tinlake.centrifuge.io/
Twitter - @centrifuge
LinkedIn - https://www.linkedin.com/company/centrifugehq/
Jeroen Offerijns:
LinkedIn - https://www.linkedin.com/in/jofferijns/
Twitter - @Offerijns
Forum - @Jeroen
Lea Schmitt:
LinkedIn - https://www.linkedin.com/in/lea-schmitt-3a17078a/
Twitter - @_schmitted
Forum - @_LS
Chat - @schmittlea
Lucas Vogelsang:
LinkedIn - https://www.linkedin.com/in/lucasvo/
Twitter - @lucasvo
Forum - @spin
Chat - @lucasvo
Website - lucasvo.com
MakerDAO:
Forum - https://forum.makerdao.com/
Chat - chat.makerdao.com
Jerry Goldfarb:
LinkedIn - https://www.linkedin.com/in/jerry-g-188a8a210/
Twitter - @JellyProducer
Forum - https://forum.makerdao.com/u/jerryag/summary
Chat - @Jerry.G
A
A
I
wanted
to
talk
about
a
couple
of
things
that
were
brought
up
last
time,
kind
of
as
follow-ups
hopefully,
and
so
first
thing
was
we
kind
of
touched
last
time
on
how
we've
been
starting
with
low
debt
ceilings
to
try
and
kind
of
scale
things
up
and
whatnot
partially,
because
this
is
sort
of
in
a
prototype
phase
so
to
speak.
I
was
curious
if
you
would
consider
from
a
technological
perspective
yourselves,
as
in
more
of
like
a
v1
or
v2
at
this
point
or
something
along
those
lines.
C
B
We
can
talk
about
it
yeah,
so
so
for
anyone
who's
new
on
the
on
the
caller
to
centrifuge
and
and
sort
of
the
focus.
What
we
want
to
talk
about
today
is
the
smart
contracts
that
enable
a
large
part
of
what
we're
doing
and
what
we're
doing
is
we
allow
sort
of
an
asset
originator
to
originate
different
loans
to
different
people
at
different
terms,
with
like
different
default
rates
and
interest
rates,
different
risk
and
then
bundle
those
into
what
we
call
pools.
B
Those
pools
have
a
have:
have
the
ability
then,
to
take
on
different
investors
that
buy
tokens
in
these
pools
to
advance
money
to
sort
of
loan,
lend
that
money
to
to
these
different
borrowers,
and
so
this
gives
you
in
in
sort
of
traditional
financial
world
is
called
called
securitization.
You
take
different
assets,
you
put
them
into
one
one
bucket
and
then
you
sort
of
issue,
a
security
that
represents
a
share
in
this
bucket
and
so
the
tin
lake
smart
contracts.
B
That's
like
a
very
high
level
description
on
the
sorry
for
the
background
noise,
but
on
the
as
an
overall
sort
of
that's,
that's
the
simplest
description
of
what
what
did
the
10x
smart
contracts
do
and
so
like?
Maybe
we
can
build
up
a
bit
around
the
questions
and
go
more
in
detail.
You
brought
up
an
interesting
point
on
like
how
what
the
maturity
is
or
like
what
the
pilot
is
like
how
we're?
How
are
we
building
this
out?
The
reality?
B
Is
we've
been
working
on
this
together
with
maker
for
over
two
years
now
we
did
first
transactions
actually
back
when
single
collateral
die
was
still
live
and
multicultural
die
hasn't
shipped
yet
and
at
the
time
there
was
no.
It
wasn't
really
collateral
maker.
But
we
already
did
like
first
transactions
just
settling
these
assets
and
die.
B
We
then
started
working
together
with
the
protocol
engineering
team
to
figure
out
a
way
how
we
onboard
this
as
collateral,
and
so
it's
an
ever-evolving
process,
but
I
think
we've
gone
over
like
a
few
iterations,
where
we're
now
at
a
point
where,
actually
from
a
technical
perspective,
this
this
setup
is
definitely
scalable.
It's
like
it's
been
audited
multiple
times
it's
been
deployed
several
times
for
different
pools
and
we're
sort
of
looking
at
okay.
How
do
we?
B
D
Yeah
hi
everybody
yeah,
like
lucas,
said
my
name
is
jerome.
I'm
an
engineer
on
the
sandridge
team
focused
really
on
the
smart
contracts
of
the
lake
and
some
of
the
things
around
it
yeah.
I
was
thinking
of
like
walking
through
a
few
things
today,
but
I'm
also
happy
to
answer
any
questions
that
everybody
has.
D
Yeah,
I
really
I
really
wanted
to
leave
it
up
to
like
the
group
here.
I
was
thinking
of
like
demoing,
a
few
things
in
our
ui
and
then
maybe
sharing
like
explaining
a
few
of
the
core
content
concepts.
I
know
there
are
some
questions
last
time
about
the
nav
feeds
in
our
contracts,
as
well
as
the
ebook
mechanism,
so
I
can
use
like
I
can
demo
and
walk
through
that.
I'm
also
happy
to
just
answer
questions
whatever
you
think
is
best.
A
D
Yeah
looking
good
great
okay,
I
will
give
like
give
a
bit
of
a
high
level
of
you.
First
of
what
thin
lake
is
and
like
how
our
pools
work,
I'm
probably
gonna,
say
some
things
that
some
people
here
already
know,
but
just
to
make
sure
they're
all
kind
of
on
the
same
page.
So
basically
thin
lake.
Is
it's
an
unchained
securitization
protocol?
So
if
you
go
to
the
clinic
apple
centrifugal
io,
you
can
see
that
we
currently
have
just
over
31
million
tpl.
D
We
have,
I
think,
10
pools
or
so
live.
So
every
one
of
these
pools
is
one
kind
of
asset
that
we're
working
with
and
it's
and
you
can
see
that,
for
example,
for
zunify.
That
was
one
of
the
pools
that
is
working
with
maker
and
it
just
got
onboarded.
I
think
a
month
ago
currently
has
1.3
million
in
in
tv
in
value.
D
If
you
click
on
this,
if
you
click
on
it,
you
can
you
go
to
the
pool
of
a
few
and
there's
a
few
metrics
here
on
top
that
I
think,
give
a
good
overview
of
what
this
pool
is
about.
So
you'll
see
the
asset
type
revenue-based
financing
in
this
case
and
you'll
see
the
underlying
maturity.
So
this
is
basically
from
the
time
that
the
loan
is
originated,
how
long
it's
going
to
take
for
this
loan
to
be
paid
off
on
average.
D
Then
you
have
the
drop
api
apy,
which
is
kind
of
the
return
for
the
senior
investors
and
I'll
explain
more
about
that
later,
and
then
we
have
the
full
value,
as
I
mentioned
before,
and
so
specifically,
this
pool
is
actually
already
operating
is
has
gone
through
the
maker
process
and
as
a
maker
fault,
it
has
a
50
million
debt
ceiling
which
they're
soon
going
to
use,
but
yeah,
that's
so
yeah.
That's
super
exciting
to
see
that
going
live
if
we
scroll
down
here.
D
This
is
what
gives,
I
think,
a
pretty
good
overview
of
kind
of
what
is
in
the
pool
right
now.
So
you
can
see.
This
is
basically
a
kind
of
a
balance
sheet
view
of
of
the
pool.
So
on
the
left
side
you
have
the
assets.
We
have
the
asset
value
at
the
top
1.3
million.
What
this
number
is
and
where
this
is
coming
from.
That's
what
I'm
going
to
focus
on
later,
but
for
now,
let's
just
say
it's
1.3
million
and
there's
six
assets
in
the
pool
on
efforts,
200
something
a
thousand
die.
D
The
average
financing
fee
of
each
of
these
assets
is
15.
So
that
means
the
borrower
on
thin
lake
is
paying
15
interest
on
a
yearly
basis
and
there's
currently
almost
no
reserve
in
the
pool,
so
that
the
reserve
is
the
money
that's
in
the
pool,
that's
not
currently
allocated
to
assets
before
I
go
on
like
if
anybody
has
any
questions
like.
Please
just
interrupt
me
or
maybe
jerry.
D
Basically,
if
there's
going
to
be
any
losses
in
this
pool,
the
thin
token
takes
these
losses
first,
so
the
drop
tokens
are
safer
and
more
protected.
This
may
also
mean
they
have
a
slightly
lower
yields,
but
are
more
stable
yields
and
more
expected
deals,
and
it's
worth
noting
also
here
that
here
you
can
see
this
minimum
of
10
percent.
This
basically
means
that
in
this
particular
pool,
the
drop
tokens
are
protected
by
at
least
10
percent
of
thin.
D
So
at
least,
if
there's
ten
percent
of
losses,
though
that
should
be
covered
by
yeah.
That's
in
specifically,
I
wanted
to
also
show
the
assets
page,
because
this
is
pretty.
I
think
it's
not
probably
the
page
that
most
people
look
at
very
often,
but
it's
quite
unique,
because
if
you
compare
this,
what
we're
trying
to
do
with
traditional
finance,
if
you
have
securitizations
these
asset
lists,
are
not
going
to
be
public
they're
not
going
to
be
in
a
fancy.
D
Ui
they're
going
to
be
in
in
long
spreadsheets,
and
you
can't
just
kind
of
see
easily
what's
going
on.
But
here
you
can.
You
know
for
any
pool
like
any.
Anybody
can
go
to
this
ui
and
just
check
out
what
are
the
assets
currently
in
this
pool.
So
you
can
see
six
assets,
certain
financing
dates,
maturity,
dates,
amounts
and
so
on,
and
if
you
click
on
them,
you
can
even
see
a
bit
more
more
information.
Yeah,
that's
on
a
very
high
level.
D
D
E
D
It's
fine.
I
can
start
with
the
continue
nav
calculation,
so
what's
pretty
crucial
for
all
of
this
to
work
is
that
we
need
to
know
what
is
the
current
value
of
the
assets
that
are
in
the
pool,
and
so
we
have
this
list
of
assets
in
the
application.
Let
me
go
here,
so
I
want
to
go
through
an
example
of
what
might
happen.
So
let's
say
we
have
a
pool
with
currently
three
assets.
D
D
The
question
is
so
you
have
this
pool
of
three
assets
with
three
assets
due
in
in
the
next
few
months.
What
is
that
currently
worth
today,
and
so
the
the
way
to
calculate
this
in
a
way
that's
just
traditionally
calculated
also
in
nutrition,
finance
is
using
what's
called
a
discounted
cash
flow
formula.
So
basically
the
first
thing
we
do
is
we
look
at
what
are
the
kind
of
risk
profiles
of
these
assets,
so
here
we're
kind
of
adjusting
the
cash
flows
for
this
risk?
D
So
yeah,
basically
we're
saying
for
each
of
these
assets
we're
adjusting
them
for,
for
what
the
risk
of
each
asset
is
and
then
what
we're
doing
is
we're
using
this
discounted
cash
flow
formula.
I
don't
think
it's
necessary
to
go
like
fully
explain
this
now,
but
basically
we're
saying
an
asset.
That's
word:
that's
due
in
a
month
from
now,
that's
quite
risky
is
worth
some
slightly
less
today.
D
We've
implemented
the
smart
contract
that
fakes.
All
of
these
asset
values
takes
off
the
maturity,
dates
all
of
the
risk
scores
and
then
per
second
is
able
to
calculate
what
is
the
current
nav
of
this
pool.
So
at
any
time
you
can
go
through.
You
can
go
to
the
thinlit
ui.
You
can
check
out
what
is
the
current
asset
value
and
that's
going
to
grow,
hopefully
over
time,
and
it's
constantly
changing
and
it's
it's
calculated
by
the
smart
contract.
D
So
as
long
as
you
trust
the
smart
contract
you're
able
to
trust
that
the
asset
value
is
at
least
calculated
correctly.
This
is
probably
quite
a
lot
of
information
again
for
some
people,
maybe
some
repetition
with
others,
maybe
a
lot
of
new
things.
Let
me
stop
quickly
again.
If
you
have
any
questions.
F
F
D
Yeah,
that's
a
really
good
question,
actually
something
we're
working
on
a
lot
quite
right
now.
So
if
you
look
at
our
current
pools,
most
of
them
the
essential
it's
a
single
s,
originator
for
the
pool,
but
actually
we're
really
focused
right
now
on
moving
to
a
model
where
there's
an
issuer,
that's
managing
the
pool
and
there
can
be
multiple
asteroids
in
this
pool.
And
the
reason
for
this
is
that
there's
many
there's
a
combination
of
reasons
for
this.
D
D
So
if
you
have
multiple
asset
originators
in
one
pool,
you're
kind
of
having
different
assets
having
different
risk
profiles,
that's
ultimately
better
and
then
it
also
aligns
very
well,
I
think,
with
the
maker
of
the
strategy
of
kind
of
having
larger
pools
that
have
that
are
kind
of
professionally
managed
and
have
multiple
estuaries
there.
So
it's
a
really
good
question.
D
Currently,
most
are
single
as
the
originators,
but
in
the
next
few
months
and
the
next
year,
you'll
see
more
and
more
pools
where
there's
one
issuer
and
multiple
asset
originators,
and
I
think
if
you
go
to
the
link
divide,
the
gig
pool.
If
I'm
not
mistaken,
it's
the
first
pool,
that's
yeah
using
this
new
concept.
F
This
is
a
hard
topic
because
I
know
the
owner
of
and
I
would
need
to
control
how
to
to
create
the
documents
how
to
meet
the
nfts
and
how
to
public
in
so
you
need
to
control
other
flow.
So
if
you
have
many
pools,
how
can
you
connect
that
to
one
sequel
to
one
single
thing,
like
there's
a
lot
of
work
here,
there's
a
lot
of
work
in
that
topic.
G
D
D
I
think
currently,
there's
there's
a
lot
of
work,
not
just
on
the
technical
side,
but
also
on
the
legal
side
that
you
need
to
do
specifically
on
the
technical
side
about
minting
nfcs.
I
would
say
that's
something
we're
very
focused
on
making
as
easy
as
possible.
We
have
a
kind
of
an
app
that
we
call
gateway,
which
is
specifically
for
ester
originators,
which
they
can
use
to
create
new
assets
and
originating
websites
and
maintenance
fees
for
them.
G
Got
it
so
if
I'm
a
pencil
manufacturer
in
panama,
it
might
be
a
little
bit
difficult.
I
mean
I
get
the
legal
size,
I
get
it.
That's
always
difficult
the
legal
structure,
okay,
cool
yeah,
so
I
guess
we're
we're
not
quite
there
yet,
but
eventually,
hopefully
it
will
be
easier
for
a
pencil
distributor
in
costa
rica
to
to
use
this
cool.
D
Yeah
for
sure-
and
I
mean
what
we
really
hope
is
that
they
might
not
necessarily
need
to
use
the
lake
themselves.
They
might
go
through.
An
issue
like
we
discussed
of
so
like
one
of
our
issuers
might
create
like
a
technology
solution.
That
kind
of
simplifies
the
process
for
them
and
then
used
to
make
this
kind
of
underlying
infrastructure,
but
as
an
easier
to
use
app
that
the
essential
pencil
manufacturer
can
use.
H
H
The
with
each
of
these
pools
has
to
value
the
the
net
asset
value
at
each
window
in
time,
and
then
you
were
kind
of
describing
how
how
you
you,
you
know,
do
re-underwrite
the
portfolio
to
come
up
with
the
value
of
the
pool.
Are
there
any
plans
to
to
kind
of
work
with,
like
a
a
third
party,
an
independent
third
party,
firm
that
that
does
that.
D
I
think
that
really
depends
on
the
on
the
pool
that
we're
talking
about,
so
that
the
nav
calculation
is
something
that
happens
on
chain
by
the
tin,
smart
contracts,
but
it's
based
on
the
parameters
that
are
set
per
asset,
so
the
value
the
risk
score
and
so
on
how
that's
set
and
who's
involved
in
setting.
That
is
a
process
that
depends
depends
on
the
pool,
and
it's
also
something
that
I
think
is
evolving.
D
I
think
right
now,
it's
often
something
that
is
is
it's
a
responsibility
of
the
the
history
leader
or
the
issuer,
but
I
think
when
probably
liara
lucas
has
some
more
color
here.
Some
people
are
working
on
kind
of
adding
independent
parties
to
this
process.
A
big
part
of
that-
and
I
don't
think
we
have
time
to
talk
about
this
today-
is
we're
working
on
on
a
new
concept
for
improving
this
further,
which
we
call
the
underwriter
token.
D
H
D
Yeah,
I
think
that's
definitely
true
and
I
think
again,
that's
another
reason
why
the
issuer
model
that
we're
trying
to
focus
on
now
makes
a
lot
of
sense
where
you
have
a
separate
issuer.
That's
managing
the
pool
that
has
multiple
edged
originators
where
indeed
you're
able
to
achieve
more
skill,
and
then
these
things
start
to
make
a
lot
more
sense
for
sure.
B
Yeah,
I
think
sort
of
thinking
about
scalability
is
really
the
like
sort
of
saying
if
we
growing
these
pools
like
setting
up
like
500
000
in
in
like
engineering
and
legal
expenses
to
pull
in
external
data
sources
for
pool.
That's
like
two
million
dollars
in
size
right
is
a
little
bit
unreasonable.
But
when,
as
you
sort
of
start
thinking
of
how
you
can
grow
this,
like
10x
100x
right
it
becomes,
it
becomes
more
relevant
and
yeah.
I
mean
I
don't
think
going
into
specifics.
Of
of
what
these
data
sources
will
look
like.
E
So
aaron
here
my
really
fascinating
stuff
trying
to
keep
up
with
with
all
these
details
as
we
go
here,
but
but
my
questions
around
you
know
there's
on
chain
and
there's
op
chain
data
here
right.
So
if
we're
looking
here
at
this
diagram
of
the
risk
adjusting
of
future
expected
cash
flows,
you
know
you
can
try
to
be
as
granular
and
informed
as
you
can
about
that
risk
adjustment
to
those
future
cash
flows.
But
but
what
about
events
where
the
off
chain
data?
E
So
maybe
the
actual
payments
of
the
mortgages
goes
out
of
sync
with
what's
represented
on
chain?
So
an
example
would
be
what,
if
the
risk
adjustment
score,
you
know
for
the
for
the
30-day
cash
flow
or
the
60-day
cash
flow
was
off
and
the
the
payments
that
were
actually
received
was
way
under.
Would
there
need
to
be
an
adjustment,
and
how
would
that
flow
through
to
to
the
drop
tokens?
You
know,
would
you
potentially
see
drop
tokens
that
are
adjusted
downward
and
how
would
that
work.
D
Yeah,
that's
a
really
good
question,
so
maybe
first
a
little
bit
of
an
overview
on
what's
unchained
versus
what's
off
chain
for
thin
lake,
and
so
what
you
can
see
on
chain
is
basically
per
asset,
all
the
data
that's
needed
for
this
nav
calculation.
So
you
can
see
the
value
the
maturity
data
risk
score.
But,
as
you
very
rightly
point
out,
this
risk
score
is
based
on
a
lot
of
other
data,
but
for
I
think
for
clear
reasons.
D
Not
all
of
this
can
be
public
information,
so
we
actually
have
a
peer-to-peer
protocol
for
kind
of
sharing
our
for
distributing
this
kind
of
off-chain
data
on
an
asset
level.
So
when
an
ester
originator
is
creating
these
assets
in
our
in
our
gateway,
app
they're
entering
a
lot
more
information,
that's
going
to
remain
off-chain
but
can
be
shared
with
some
parties
kind
of
what
goes
into
the
risk
score,
and
so
this
information,
for
example,
is
shared
with
maker
with
a
maker,
rpa
team.
D
D
It
works
it's
basically
on
an
asset
level
that
they,
like
smart
contracts,
support
a
way
to
update
either
the
value
of
that
the
of
the
of
the
underlying
assets
to
the
collateral
or
to
update
the
risk
score,
and
so,
if
the
the
risk
core
of
an
asset
would
update,
this
would
be
reflected
in
a
fee.
So
if
the
asset
becomes
higher
risk,
yeah
the
asset
value
would
go
down.
I
hope
that
answers
your
question.
E
Yeah
and
then
I
guess
just
carrying
it
carrying
it
a
step
further,
if
you
did
have
a
downward
adjustment.
You
know,
I
guess,
thinking
of
other
maker
collateralized
debt
positions
right.
We
have
real
time
price
updates
on
chain.
If
your
collateral's
ether,
for
example,
I
could
foresee
an
event
here
where
you'd
have
a
lag
of
you
know.
E
Let's
say
you
needed
a
downward
adjustment,
but
there
was
a
question
mark
around
how
severe
that
downward
adjustment
needed
to
be,
and
then
my
next
question
would
be
around
you
know
over
collateralization,
so
how
much
dye
is
being
minted
against
these
assets
relative
to
to
the
value
of
the
underlying
collateral,
and
is
there
a
buffer
in
there?
For
for
this?
This
sort
of
you
know
potential
problems
right
that
that
could
come
about.
D
Yeah
for
sure,
so,
yeah
you're
absolutely
right
that
there
is
a
bit
of
a
it's
not
the
same
as
on
chain
crypto
collateral
in
terms
of
price
updates.
It
takes
a
little
bit
more
time,
but
there's
also
like
pretty
significant
protections
built
in
so
there's,
basically,
three
levels
of
over
claritablization
in
the
the
lake
protocol
for
the
maker
fault.
D
So
on
the
most
granular
levels,
every
asset
has
an
advance
rate
and
so
against
the
value.
Only
that's
in
ninety
percent
of
the
of
the
value
can
be
can
be
borrowed.
So
that's
the
first
kind
of
level
of
protection
already.
Then
we
have
this
senior
junior
trans
structure,
where
the
thin
token
is
already
protecting
the
drop
token
for
literally
ten
percent
of
the
losses
and
so
drop
or
so
maker
is
only
investing.
D
It
drops
our
makers
always
protected
by
this
thin
buffer
and
then,
thirdly,
what
specifically
maker
is,
or
the
maker
faults,
it's
receiving
an
additional
level
of
globalization.
So
I
don't
know
the
exact
numbers
of
the
top
of
my
head,
but
I
think
example
for
new
so
far,
the
largest
the
lake
pool
with
maker
right
now.
I'm
pretty
sure
that
there's
105
percent
over
collateralization,
so
basically
when
die,
is
drawn
actually
105
of
drop
gets
moved
into
the
maker
folders
collateral
collateral.
D
E
D
That
depends
per
pool,
I
think,
for
all
pools.
The
asteroids,
nato
or
the
issuer
is
holding
some
dim
tokens,
if
not
the
majority
of
it,
and
then
sometimes
they're
they're,
also
able
to
there's
other
investors
are
able
to
invest
into
tin.
But
it's
not
something
where
there's
a
lot
of
right
now.
As
a
retail
investor,
it's
you
can
only
sign
up
for
the
drop
token
by
default,
because
the
investing
in
the
drop
token
requires
a
lot
of
investigative.
D
I'm
sorry
requires
a
lot
more
understanding
of
the
underwriting
assets,
my
risk
profile
so
yeah,
it's
not
it's
not
as
public,
but
by
default.
It's
at
the
very
least
as
to
register,
is
investing
some
part
in
at
the
end,
so
they
basically
have
a
first
loss
position
to
clarify
tin.
It's
basically
in
most
cases
it's
it's
a
it's
an
accuracy
tranche
and
rather
than
a
messenger
ranch,
except
in
some
pools
where
there's
also
some
kind
of
off
chain
equities
and
not
all
of
equity
is
inattentive.
E
Okay
and
then
so
that
also
gets
into
the
skin
of
the
game
skin
in
the
game
question
around
the
asset,
originators
so
you're
saying
that
they're
also
holding
the
the
10
tranches
there,
so
they
do
have
skin
in
the
game
right.
If,
if
the
the
those
the
investments
are
not
the
payers
are
not
meeting
their
obligations,
10
the
10
holders
being
the
asset
originators,
there's
a
connection
there,
yeah.
D
Yeah
100-
and
I
think
I
think,
for
all
of
the
maker,
if
all
of
the
pools
that
are
on
board
of
this
maker,
clutter
right
now,
there's
an
additional
confidence
that
they
need
to
hold
the
percentage
of
a
certain
percentage
of
the
tin
and
actually
we're
working
on
an
update
to
the
ui
that
we're
also
going
to
be
showing
this
better
in
the
108.
Ui.
So
you'd
be
able
to
see
how
much
stint
that
the
actual
asteroid
center
is
holding.
B
B
J
I
J
Yeah,
so
so
the
discount
rate
I
mean
there
are
two
approaches
or
textbook
approaches:
one
is
benchmark
rate
from
reflecting
cost
of
capital,
with
a
comparable
amount
of
risk.
Basically
right
this.
This
is
what
we're,
basically
following
so
trying
to
find
the
expected
average
financing
fee
of
the
portfolio
risk
adjusted
to
come
to
the
cost
of
capital
to
discount
this.
The
other
would
be
like
what
you
also
find
then
is
using
a
risk
free
rate.
J
This
is
the
textbook
approach.
Usually
this
doesn't
make
too
much
sense
in
our
in
our
case
here,
so
we
use
the
basically
a
benchmark
rate
that
reflects
comparable
risk,
adjusted
rate.
Basically
for
this,
for
these
types
of
assets
right
so
right,
this
discount
rate
is
always
like
you
have
this
opportunity,
thought
behind
it
right,
you
think
like
okay,
like
I
discount
it,
because
I
could
also
do
something
else
with
my
money,
so
I
I
want
a
benchmark
benchmark
right
to
see
what
it's
worth
today,.
I
Yeah
so
totally
understood,
so
I
I
guess
I
just
meant
like
for
your
particular
methodology
right,
like
on
a
new
silver
pool
like
how.
How
do
you
guys
at
centrifuge,
go
about
figuring
out
what
the
benchmark
is
and
like?
If,
let's
say
you
know
broader,
like
the
spread
above
the
risk-free
rate
and
the
broader
markets
moves?
Is
that
reflected
in
the
discount
rate?
Or
do
you
just
try
to
choose
a
benchmark
that
gauges
the
opportunity
cost
of
capital
at
the
beginning
of
the
pool
and
then
kind
of
just
go
with
that?
Does
that
make
sense.
J
J
So
if
the
risk
of
the
underlying
portfolio
would
would
change
significantly,
you
would
need
to
change
the
discount
rate
if
interest
rates
change
and
that's
the
financing
fee,
for
example,
right,
like
financing
fees
drop
for
drop
by
two
percent
yeah,
then
obviously
also
the
the
discount
rate
would
need
to
be
adjusted
at
some
point
to
reflect
this
new.
This
new
benchmark.
J
Yeah
tin
buffer,
I
mean
we
roughly,
it
depends
obviously
on
the
on
the
risk
of
the
underlying
of
the
underlying
asset
right.
This
is
like
rule
of
thumb,
the
the
riskier
the
asset,
the
higher
the
tin
risk
buffer
is-
and
this
can
depend
like
this
can
be
credit
risk.
This
can
also
be
can
also
be
liquidity,
risk
right,
like
all
types
of
risks
that
go
in
there
and
then
the
higher
the
risk,
the
larger,
the
tin
buffer.
J
I
think
it's
fair
to
say
like
this
is
not
a
scientific
approach
right.
We
don't
calculate
a
tin
risk
buffer
of
11.782
because
there's
some
water
underlying
it,
but
so
far
we
usually
go,
which
is
in
line
with
first
loss
pieces
in
securitizations
and
other
structures
that
you
have
a
junior
piece
of
10
to
15,
usually,
and
that's
what
we're
using
so
far.
D
The
difference
to
your
specific
question
between
author
and
existing
testnets,
like
flint
and
amber,
is
that,
as
is
kuzama,
altair
is
going
to
be
a
canary
mainnet.
So
this
means
there
will
actually
be
real
value
on
this
network,
and
you
can
see
with
this
happening
with
polkadot
and
and
the
intention
behind
kasama
as
well.
D
We're
trying
to
create
a
network
that
allows
them
to
test
new
features
and
to
experiment
with
new
new
ideas
with
some
value,
but
still
keep
the
main
network
polkadots
in
our
case
interface
chain,
more
stable
and
more
secure,
so
yeah
the
difference
between
all
their
and
emery
and
finn's.
Emer
and
flint
are
just
test
nets.
There's
no
real
value
there,
but
altair
has
actual
military
value.
D
Alrighty,
if
there
are
no
more
questions,
we're
also
going
through
time
quite
quickly,
maybe
I
can
give
take
five
minutes
to
walk
through
the
epoch
model
still
and
then
do
our
last
round
of
questions.
So
this
is
a
bit
of
a
chart,
but
I
have
it's
the
best
visualization
that
I've
found,
so
our
thin
leg
pools
are
what
we
call
revolving
pools.
So
initially,
actually
we
launched
an
older
version
of
thin
lake
which
only
supported
static
tools.
So
what
this
means
that
you
had
at
some
point,
the
pool
would
start.
D
Investors
would
submit
money
and
either
invest
money
into
the
pool
and
then
asked
originator
would
finance
assets
using
that
money
and
then
in
the
end
you
pay
them
and
then
the
investors
would
redeem
your
capital
and
then
the
pool
would
be
closed
again.
But
this
was
very
inefficient
because
she,
you
couldn't
be
able
you
couldn't
easily
like
finance
new
assets.
D
You
always
have
to
spin
up
a
new
pool,
and
so
we
started
working
on
creating
a
new
version
of
the
lake
that
supports
revolving
tools
and
that's
what's
live
today,
and
what
a
lot
of
our
pools
right
now
are.
Revolving
pools,
the
nav
calculation
is
a
crucial
part
for
this
to
function,
and
the
other
part
of
this
is
the
epoch
model.
D
So,
basically,
at
any
time
that
the
link
pool
is
in
a
certain
epoch,
most
pools
right
now.
The
e-books
take
about
one
day
in
that
e-book
any
investor,
any
any
investor
can
submit
orders.
They
can
say
I
want
to
invest
this
amount.
I
want
to
redeem
this
amount
and
they
won't
immediately
be
fulfilled.
They'll
basically
be
collected
during
the
epoch
and,
at
the
same
time,
esther
engineers
can
finance
assets
as
well
as
make
repayments,
and
these
repayments
aren't
the
money
that's
coming
in
from
these
repayments.
D
D
So
at
any
time,
if
a
drop
investor
or
and
investor
would
want
to
redeem
any
of
the
capital,
this
should
take
priority
over
the
estrogen
being
able
to
finance
new
assets.
So,
at
the
time
of
the
epoch,
when
it
closes
our
contracts
calculate
what
amount
of
the
orders
can
be
fulfilled.
So
maybe
part
of
the.
So
I
can
actually
have
an
example
here.
D
So
this
is
kind
of
an
ongoing
process
every
day.
There's
a
new
epoch
also
handled
entirely
very
smart
contracts
and
yeah,
that's
kind
of
what
enables
these
schools
to
be
revolving
and
what
enables
yeah
investors
to
be
constantly
participating
and
certain
areas
to
be
constantly
originating
new
assets
and
so
on.
G
D
E
C
C
Yeah,
maybe
it's
because
yeah
I
mean
overall,
it's
like
it's
like
an
open
system.
Theoretically,
anybody
could
spin
up
a
pool
or
launch
a
pumpkin
like
and
yeah.
I
mean
it
it
basically,
what
what
they
would
need
to
do
is
like,
like,
like
map
their
their
business
model,
basically
or
like
basically
a
map,
their
business
model
or
to
entertain
like
they
in
order
to
launch
a
pool
like
what
this
is
more
or
less.
C
These
steps
that
we
went
through
at
the
beginning,
like
where
we
look
at
okay,
what
how
is
the
asset
originator
or
how
is
an
issue
structuring
their
pool?
What
is
the,
what
is
the
asset
type
that
they
seek
to
finance?
How?
How
would
it
work
with
the
tin
leg?
Mechanics?
What
is
the
discount
rate?
What
are
the
interest
rate
etc?
D
F
D
D
Yeah,
so
in
this
case,
all
of
the
capital
would
be
would
be
unallocated
and
would
go
into
the
pool
reserve
and
basically
there
wouldn't
be
anything
technical
that
would
block
the
pool
from
functioning,
but
you'll
quickly
start
to
see
that
the
yield
is
going
to
go
down.
Obviously,
because
no
there's
no
interest
agreeing
on
on
another
allocated
capital
on
a
reserve.
So
I
I
would
imagine
that
all
of
the
investors
in
the
pool,
including
maker,
would
see
their
yield,
dropping
and
would
start
to
submit
redeem
orders.
D
E
Can
can
I
ask
a
question
on
and
sorry
if
it
was
already
covered?
I
joined
a
few
minutes
late,
but
the
relationship
here
with
the
nf
nft
tokens
the
non-fungible
tokens
specifically
what
those
represent
in
terms
of
off
chain
assets
and
then
the
erc
20
tokens
that
are
minted
against
those
which
it
sounds
like
that's
the
drop
in
the
tin
a
little
bit
around
that
process
and
then,
specifically
on
the
drop
in
the
tin.
E
D
Yeah
for
sure
so
yeah,
the
any
asset
that's
in
the
pool,
is
linked
to
an
nfd.
That's
representing
the
underlying
collateral,
and
so
this
nft
actually
is
is
anchored
on
our
own
chain
and
contains.
Basically,
it
contains
a
link
to
the
off
chain
data.
That's
representing
collateral,
so
yeah,
the
any
and
any
asset
in
the
pool
is
linked
to
an
nfc.
And
then
the
pool
makes
up
this
kind
of
the
set
of
the
nfts
and
then
the
drop
in
the
tin
tokens
are.
I
D
G
H
Can
you
just
talk
a
little
bit
about
like
the
process
of
of
I
mean,
like?
Does
an
asset
originator
kind
of
have
have
their
own
ui
on
tin
lake?
So
they
can?
Is
it
the
asset,
originators
responsibility
to
to
kind
of
monitor
incoming
investments
and
and
as
well
as
redemptions
and
then
and
then
maybe
like
a
little
bit
of
color
on
on
kind
of
the
cash
management
process?
D
Yeah
for
sure
so
yeah
in
terms
of
managing
the
epochs,
so
yeah
epochs,
although
they
run
on
a
daily
basis,
they're
not
automatically
closed.
Someone
needs
to
do
that.
Unfortunately,
because
of
ethereum
gas
feeds,
that's
that's
not
very
cheap,
so
yeah.
That
needs
to
be
a
proactive
step.
In
theory.
I
think
it's
worth
noting
that
this
is
kind
of
permissionless.
D
Anyone
could
just
close
the
epoch
for
any
pool,
but
it's
going
to
cost
you
some
money
and
that's
probably
the
biggest
holding
like
biggest
reason
to
not
do
that
yeah!
It's
I
I
would
say
it
is
the
responsibility
of
the
issuer
of
any
pool
to
actually
manage
that,
and
they
do
indeed
have
their
own
ui
within
the
lake
to
do
that,
and
they
also
get
kind
of
proactive
notifications
about
what's
going
on
in
their
pool.
D
So
they
get
notifications
if
there's
new
orders
coming
in
or
if
they've
not
been
fulfilled
for
a
while
and
it's
up
to
them
to
manage
that
they're
also
able
to
kind
of
signal
their
their
interesting
capital.
I
would
say
so.
Maybe
you
saw
recently,
I
think,
a
month
ago,
or
so
we
rolled
out
a
new
feature
in
the
ui,
where
you
can
see
the
investment
capacity
for
any
pools.
You
can
basically
see
how
much
can
currently
be
invested
into
the
drop
token
of
this
pool.
D
H
Okay,
great
and
and
just
one
quick
follow-up
on
that,
can
you
talk
a
little
bit
about
like
the
on-ramp
and
the
off-ramp
like
if,
if
die,
if
an
lp
is
lending
dye
into
a
pool?
Is
there
kind
of
like
a
standardized,
on-ramp
and
off-ramp
to
get
that
die
into
you
know?
I
I
imagine,
like
a
bank
account
of
an
asset
originator.
D
I
think
what
most
estuarines
are
doing
is
they're,
first
converting
the
dye
into
the
offer,
I'm
first
converting
the
dye
into
usdc,
using
either
the
psm
or
a
curve
and
then
yeah
trying
to
use
it,
transferring
that
that
ucc
into
into
fiat,
but
that's
not
something
that
we're
necessarily
kind
of
telling
them
how
to
do
it's,
just
something
that
seems
to
work
best
for
most
pools.
I
think
yeah,
I
don't
know,
do
you
have
anything
to
add
there
or
lucas.
D
Think
this
is
also
something
where
yeah
the
issue
or
model.
Where
will
help,
I
think,
over
time
there
will
likely
be
polls
where
the
escher
is
handling
all
of
this,
and
actually
the
estuary
center
is
entirely
just
dealing
in
fiat
and
doesn't
really
have
to
be
involved
with
the
written
script,
though,
which
will?