Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
DeFAI on Arbitrum - Open Call
Recorded: Feb. 6, 2025 Duration: 1:17:01Space Recording
Short Summary
The discussion highlighted several key developments in the crypto space, including the launch of new projects like CryptX's decentralized indexes and CookieDAO's Agent Cookie. There are ongoing efforts to integrate AI into DeFi, such as CryptX's yield optimization strategies and a cross-chain yield optimizer. Partnerships are forming, like Smoothie Phone's collaboration with Aura, and grant programs are supporting AI projects on Arbitrum. Additionally, upcoming token launches and IDOs are planned, indicating continued growth and innovation in the ecosystem.
Full Transcription
Hello, hello, hello, guys.
Appreciate you all for joining.
We'll be starting momentarily.
I'm just waiting for another co-host of mine and all the people to show up.
had and some good collaborations that can be
done.
Should we start
bringing people on stage?
Yes. Before we do,
just to get out of the way,
for those of you who don't know, I'm sure most people do already,
there is a grant program right now
very specifically allocating
money and funds to people
building with AI on Arbitrum, whether it's
DeFi adjacent or not. If you're
innovating with AI, we would love for
you guys to take a look at the pinned tweet.
Tons of funding
is available. We just want to make sure people are
building the coolest shit and
people are seeing it.
All right. Cool. We had
CryptX already up here.
Guys, if you want, maybe feel free to
pitch what you're doing already on Arbitrum
and maybe what your
view is on AI and maybe how you plan on
integrating it or if you already have it integrated.
Let us know what that looks like.
Absolutely. What's up,
Arthur? It's Joe. How you doing, Joe?
I'm doing good, man. Doing good. How are you?
Doing great, man. Excited.
So thank you, first and
foremost. Thank you for having me and
hello, everybody.
Just want to share a little bit about what we're
working on. CryptX is currently building
all different types
of fully decentralized indexes
and perpetual markets
on multiple L2s, including
Arbitrum. All of our
oracles are powered by chain link data
streams and price feeds.
And, you know, we've been
we've been working on a lot of really cool things,
you know, in the Arbitrum space for some time.
Specifically, we deployed a meme index
that takes five
meme feeds, Doge,
Wiff, Shiba.
We put Ethan there to
offset it
amongst a couple others.
And essentially, basically, like that index is aggregated
and people can trade an index of
memes in perpetual format.
Have some other really cool indexes
coming to Arbitrum.
We're looking into doing
an Arbitrum DeFi index
where it's specifically
DeFi projects within the Arbitrum ecosystem
that we kind of want to
index and have like a one-click option
where you just mint one token.
It automatically provides exposure to those indexes.
But more specifically for this call,
I'm working on a little bit of a new project
and kind of want to speak with
some of the Arbitrum teams
on ways that we can merge AI
into our existing stack.
Learn a little bit more about
what some of the teams are doing
specifically around
yield optimization strategies
and how AI is working to power those.
I have a really, really cool idea.
And hopefully after this call,
we can connect with some teams
and maybe chat about it
and work together.
Can we learn a little bit more
about the idea
or is it too early?
Well, it's a little bit early,
but specifically,
you know, what we're seeing now,
especially with what
Offchain released the other day
with regards to interoperability
across the Ethereum space,
we think that this is going to be
something that continues
where Ethereum essentially becomes
an interoperable network
amongst L2s
where you can bridge assets
seamlessly across each L2.
If so, you know,
we think that there's
an incredible opportunity
for things that are like
actively scanning
for yield pools
across multiple L2 networks.
And as those are scanning,
you know,
you can essentially just deploy,
imagine like 100 USDC
that's just auto-deployed
into, you know,
different pools and whatnot,
always offering users
the best yield
across multiple L2 networks
seamlessly.
That's awesome.
I think that's one of the things
that we had outlined
in our original Trailblazer blog post
as like almost a request,
a request for something,
request for people to build.
Really appreciate that
you're doing that
and also appreciate
that you're paying attention
to the big Interop announcement
that we had yesterday.
Off-chain is definitely
one of the best,
in my personal opinion,
one of the best in the space.
I mean, Hunter can tell you guys,
you know,
there's a lot of synergies
between the teams,
you know,
both on-chain and off.
I mean,
we've known Hunter
for many, many years.
We know a lot of the,
you know,
off-chain guys
for many, many years
and just really blessed
to have the opportunity
to build alongside you
and watch what you guys
are doing
and, you know,
we have a really cool
yield type of index
that we're in the process
of building
but, you know,
me trying to be
a forward-thinking founder,
it's like,
well, how can we take this,
you know,
from zero to 100,
you know?
And I think maybe
if we have the opportunity
to connect with some teams
that are specifically
working with AI,
you know,
through Arbitron,
I'm sure we can come up
with something really cool,
a structure that's really cool
and our DMs are open.
You know,
we're here
and we're building
and I look forward
for the opportunity.
Yeah.
Thanks for coming on, man.
Please shoot me a DM.
Shoot me a DM after this
and my DMs should be open
and we can discuss ways
that we can be helpful.
Absolutely.
Thanks, John.
Yeah.
Appreciate it.
All right,
next up,
let's bring Tachi.
You want to talk about
some of the things you've been,
what have you done this week?
What have we done this week?
I'm stealing that.
I'm stealing that.
Hi, everyone.
I'm Tachi.
I'm the co-founder
and CEO of Playgrounds
and we're the creators
of ARC,
the ARC ecosystem
and the RIG Rust framework
for building AI applications.
A quick rundown
on, you know,
for those who might not be aware,
RIG is our open source
Rust library
for building
and deploying
lightweight,
scalable performance
AI applications
and that can perform actions
both off-chain
and on-chain as well.
And our goal with RIG
is to essentially produce
the highest quality
framework
with the right components
that allows anyone
to build applications,
practical,
real-world applications,
whether, you know,
if it's looking
into the DeFi space,
performing actions on-chain
that, you know,
simplifies what you can do
or if you're looking
at, you know,
other things
in the Web2 space
from e-commerce
to robotics
to autonomous systems,
you know,
just whatever, you know,
you can imagine
that AI can facilitate,
our goal is to essentially
enable that
with our Rust library.
And ARC itself,
you can think of it
as an open, tokenized layer
that incentivizes builders
to experiment,
build,
share their creations
and push the boundaries
of AI.
And so the way I like
to describe it
is ARC can be seen
as a magnifying glass
or a lens
and the collective intelligence
of developers
and non-developers
are light,
like sunlight.
And our goal with ARC
is to essentially focus
that light
towards a singular point,
which is the RIG library
and who knows
what could be
created
once you have
that concentrated effort
on a single place.
And so, you know,
ARC is about
two months old right now
and we've seen
an amazing growth
of developers
building all sorts
of applications,
whether it's gaming,
whether it's DeFi,
whether it's
in the medical industry.
There's a lot of really
interesting applications
being built
within the ARC ecosystem
and it's all powered
by RIG.
And what we're doing
in our bedroom
is essentially,
you know,
supercharging that
by providing
an EVM-enabled kit
that allows developers
to develop
these agents
and applications
that can perform
actions natively
on our bedroom.
Because we really think
that, you know,
with the combination
of AI agents
and networks
like our bedroom,
we can significantly
lower the barriers
of entry
when it comes
to adoption
of DeFi
and other
blockchain-based
applications.
So that's what we're
doing.
We've got a lot
of really cool stuff
cooked up
in the next few days
and I'm happy
to be here.
And my co-founder
is also here as well,
Thierry.
Hey, Thierry.
Thanks, Tachi.
Yeah,
to give you guys
a little context
on, you know,
how we came
to be working
with Playgrounds,
we sort of realized
that there's only
one Rust-based
agent framework
and there's only
one EVM chain
on Ethereum
that also supports
Rust,
which is Arbitrum
through Stylus,
which is the
co-equal
Wasm virtual machine
we've added
to Arbitrum chains.
So needless to say,
like both of our teams
are Rust maxis.
We got on
with Thierry
and Tachi
and some of our,
you know,
more Rust-heavy
product managers
and they jammed
super hard.
It was really great
to see
and we're excited
to work
with any teams
that are going
through ARC's
Handshake program.
Teams that are
going through
ARC's Handshake program
that do use
Arbitrum or Arbitrum
Stylus are also
potentially eligible
for Arbitrum grants.
So just throwing
that out there.
But yeah,
really excited
to keep working
on that agent kit.
I think we might have
leaked a little bit
of Alpha there,
but that's okay.
Yeah.
Any last words,
Tachi?
Alpha is good.
Alpha is good.
I would say
keep an eye out
and we've got
some really exciting
stuff coming up.
Something in the next
few hours.
Keep an eye out.
Oh, okay.
Yes, sir.
Nice.
Notifications.
Guys,
what is it?
Notifications on?
Yeah.
Bell icon?
Keep a notice on.
Thanks, guys.
Sweet.
Thank you so much,
Tachi.
By the way,
people who come up
and speak,
feel free to stay up
if you want.
I figure we all
make some intros
and then maybe
we can just jump
into any random
conversation that's
DeFi-AI-related.
Kristen,
I want to make sure
you get a chance.
You joined right after
Tachi.
Feel free to go ahead
and intro yourself
and let us know
what you're doing.
Happy to be here.
So I'm Christina.
I'm the CMO of Cookie3
and core contributor
to CookieDAO.
So most people
know CookieDAO
from cookie.fun
and recently released
Agent Cookie.
And this is basically
what we've been doing
like the past two weeks.
We've been working
on Agent Cookie
and right now
we're polishing him
to make him ready.
And of course,
right now,
the lock terminal
is view only,
but we're planning
to release it
as an interactive terminal
sort of like AIXBT
in the coming weeks.
So this is really exciting
and we've been grinding,
working on that
the whole week really.
So can you tell us
a little bit,
Cookie Fun
or Cookie3
for anyone who doesn't know
is probably the best place
to go
engagement metrics,
token and market metrics
related to AI agents.
Any sort of data
that you want
related to AI agents,
I would say Cookie Fun
is like the place to go.
And they obviously
have their own agent
that they've been working on.
Can you tell us
a little bit about how,
can you compare
and contrast that
to AIXBT?
Well, yes.
well,
the biggest difference
is so
there is two things.
Cookie Agent,
I would say
it's similar
to AIXBT,
but it's going
to be cheaper
and then
we believe
it's based
on like
more accurate data.
And then Cookie Fun
is more like
an index
for humans
to use.
basically,
you know,
you can go,
I would compare
Cookie Fun
to more like
CoinMarketCap
or CoinGecko
for AI agents
because you can just
go and see
data on any agent.
But in contrast
to like,
you know,
CoinGecko
and CoinMarketCap,
it's not only
like price,
market cap,
trading volume
and stuff like this.
It's also
on-chain data
for those
who have premiums.
basically,
the holder retention,
how much people
are holding,
so sort of like
the holder spread,
but also my favorite
part,
which is the social data,
which is mindshare.
how much of the
entire crypto Twitter
conversations
this agent
is like taking,
as well as
smart followers,
so account that
we flagged
as valuable
real human account
that have some
influence in this space.
how many of these
people are following
and engaging
with these
agents'
Twitters?
And basically,
what we found,
it's really
that there is
a correlation
between mindshare
and price.
if you see
a rise in mindshare,
usually it's reflected
in the rise in price
and vice versa.
A drop in mindshare
is usually followed
by a drop in price.
for me,
that's like,
that was really
interesting to see.
Yeah,
I've noticed,
I've noticed the same
thing and
it's really hard
in my opinion.
It's hard for me
to explain
or understand
that these things
are related,
but if you look
at Cookie Fun,
if you look at the data,
like,
they almost certainly
are.
There's certain agents,
I won't name names,
that have gotten
less interesting
than they were
in the beginning
and the price
has also gone down
and I think
that there's a little
bit of a reflexive loop
where when
the price
is going down,
the agent also
gets less funny.
But you don't know,
you don't know
what causes what
exactly.
is it less funny
because
your bags
are down
or is it less funny
because
it's just
less funny?
Yeah,
I don't know,
it's sort of just
a random thought.
You know,
I guess
it's also like
something cannot
be funny
like always.
You get bored
with a certain
time of humor
and especially
with like
how quickly
crypto moves
from one thing
to another.
Like things
are relevant
for let's say
like a week
or two.
If you stay
relevant for a month
like wow,
you're doing
something really,
really well
which is why
I believe
like we're going
to see more
utility agents
like you know
AIXBT
and hopefully
the agent
cookie
as well as
I'm really
looking forward
to seeing
you know
deaf AI agents
really popping up
because I feel
like their
utility is going
to be really
good.
Well,
we're actually
like with
you guys
with Arbitrum
we're hosting
the cookie
deaf AI track
hackathon right
now and Arbitrum
has got their
own track
and today
we open
submissions
and I've been
looking at
projects that
are submitting
and you know
some really
cool stuff
is being built
so there is
definitely
potential there
into more
on-chain utility
to AI agents
rather than
them just being
you know
either data
aggregators
and letting
you know
about data
or just being
funny on
crypto Twitter
like there is
so much
potential for
this utility
and I'm
really looking
forward to it.
How's our way?
Yeah, go ahead
Hunter.
No, I was
literally just
going to say
definitely looking
forward to that.
Also, I was
going to say
can't you
just add an
extra column
to all the
data that you
guys currently
have on
cookie.
And just
have a column
just be like
fun percentage
or funny
like funny
percentage.
That way we
can like
measure you
know like
how funny
an account
is and
you know
as it drops
maybe just
correlate that
with the price.
Yeah, that's
a great idea.
I would have
to like speak
to our like
product team
because it's
like how
do you measure
humor?
Like humor
is so subjective
like how
would you measure
humor?
That's a tough
one.
With tokens?
I will say
there's a difference
between like
well actually
that is a good
point, right?
I mean
Doge is pretty
high up there.
I'm not going
to lie to you.
I'd be lying
to you if I
told you I
knew how to
do it.
But yes,
I'm very
excited obviously
for that
hackathon.
Also,
Arbitrum coins
are supported
on cookie.fun
which is great.
There will be
more of a
deeper integration
some point
soon.
Just definitely
keep an eye
out for that
especially if
you're building
with Arbitrum
you're launching
your token
there.
You want to
make sure
you keep an
eye on that
when it launches.
Yeah,
and definitely
be on the
lookout for
agents that
come out of
hackathon
that we're
hosting together.
Super excited
for that.
Sweet.
Thank you so
much.
Feel free to
stay up if
you can.
If not,
no worries.
Smoothie phone.
We saw you
come next and
they'll be at
Raz right
after.
Smoothie,
how are you
doing?
Hey,
how's it
going guys?
My name is
Simple Farmer.
I'm the founder
of Smoothie
Phone and I'm
an alcoholic.
sorry.
I guess that
is the wrong
thing.
Oh my god.
Wrong
meeting.
Sorry,
sorry,
sorry.
I'm just
kidding guys.
Just a little
humor.
Love it.
Love it.
I'm glad to
hear the
cookie is
being polished,
the cookie
agent.
And I was
really glad to
hear about
Arbitrum's
intent engine,
pretty much
everything that
is cross-chain
and anything
that bridges
liquidity more
between the
chains,
between the
L2s and
of course the
L3s I'm
super excited
about.
So I was
really glad to
see that
Offchain Labs
is putting
in work.
But I
never had
any doubt
about that.
In regards
to Smoothie
Phone,
our DeFi
integration is
based on
our latest
product is
called the
DPP.
It's the
decentralized
parlay
platform.
It essentially
allows you to
create multi-leg
wagers or
bets on
sports and
now also
crypto
prices,
crypto price
predictions as
well as real
world events.
Our odds are
being sourced
from
Fandle,
DraftKings,
and now
as of today
Polymarket.
So we're
actually the
first decentralized
parlay token
allowing users
to create
parlay bets
on polymarket
pools.
The DeFi
integration comes
into play
with the way
the games get
resolved,
the way that
each parlay is
graded as a
win or a
loss.
Previously, we
were using the
ELISA framework
in conjunction
with EAI.
However, we've
now shifted,
now we're
using Aura.
They specialize
in AI
oracles, so
we've decided to
work with them.
We'll be
announcing our
partnership shortly
and they're
also a
fantastic team.
Been super
helpful.
Shout out to
Aura.
I don't know if
he's still
around.
Oh, yeah,
that's awesome.
Yeah, that's
exactly who I've
been talking to.
Also, shout out
to Raz, long
time no see
I haven't talked
to you in a
while either.
Beautiful, dude.
Again, if
anyone wants to
get in touch
with Smooth
Phone, Christina,
Tachi as well,
feel free to DM
them directly or
feel free to DM
me even.
I'd be happy to
connect everyone.
Movi, are you
building any
parlays for the
Super Bowl on
Sunday?
Yeah, I mean,
there's actually a
lot of parlay
bets have come
in, so we've
gotten close to
25K in
total betting
volume and
at one point
over 100K in
total liquidity
deposited in
the DPP.
So the DPP
stands for
decentralized
parlay platform.
It's essentially
like decentralized
sportsbook, like
decentralized
FanDuel.
And so users
can go in there
and deposit
liquidity.
And then if
other users
come in and
place bets,
if those users
lose, that yield
gets distributed
to the liquidity
pool, right?
So it's kind of
like staking.
That being said,
we have all
kinds of player
prop bets as
well as Super
Bowl bets.
And all the
top 50 pools
and soon top
100 pools you
see on Poly
Market, we're
live feeding those
odds onto the
DPP.
So then you can
create a parlay
pretty much like
mixing like a
political decision
with a crypto
price with the
winner of the
Super Bowl.
And yeah,
we've had close
to $5,000 in
USD value of
SP bets placed
in regards to
the Super Bowl.
So yeah, we're
definitely looking
forward to this
Sunday.
These kind of
events, I guess,
are really huge
catalysts for
parlay platforms.
Okay, but what
parlay are you
putting together
for the Super
Bowl?
Yeah, I've
got Chiefs and
Mahomes,
over on
TVs.
There we go.
There we go.
You know they
say when the
Eagles win or
when a Philly
team wins the
Super Bowl or
World Series,
the economy goes
into a recession.
Real fact.
For me, the
most important
thing is that
all New
England teams
don't win
sporting events
because I can't
handle the fans.
With you there.
All right.
Thanks.
Thanks a lot,
Smovey.
Really appreciate
Decentralized
parlay protocol.
Check it out.
Built with
Arbitrum Stylus.
Next up, we
have friend of the
show, Raz.
Welcome, Raz.
Hey, how is it
going?
Shout out to
Super Farmer, too.
How is it going?
How is it going?
First of all, I
just also want to
apologize.
Raz, I think, is
staying up really
late for this one.
Let's just all give
some hearts in the
chat for Raz.
All good, all good.
This is a very rare
occurrence where I
still awake, right?
So we're still
middling in something
related to our
website, right?
And then, oh, it's
2 p.m.
Sorry, it's 2 a.m.
already and
the Twitter
space is live.
So, oh, okay.
Might as well join
in there.
I thought it was
around like 2.30
in my place, my
local place, 2.30
a.m.
So, yeah, yeah.
Still awake.
Well, then, dude.
Listen, you know.
You're going to let
us know, man.
What are you building?
Sure.
Sure.
Very quick
introduction, right?
So, I think
around two months
ago, I think
in the 9th of
December, right?
I think
I think
got it, got it.
Hello, can you
can you hear me?
Yep, you're good.
Yeah.
Okay, got it.
So, yeah.
So, just
just
probably just
quick
introduction
about
what I'm
building, right?
around two months
ago,
on the 9th of
December, right?
So, we
able to deliver
the first
AI agent
that able
to facilitate
swap their
group of Twitter,
right?
So, the
name of
the AI agent
is Amy
Swap,
right?
Amy Swap
went
very, very
well,
right?
Went
quite viral
in a
mini-viral.
And it
went too
well
until
one month
after the
release,
Twitter actually
suspend
the account
because,
you know,
they
deem it like,
they deem
the account
is too
spammy,
right?
Because people
are just,
you know,
just keep
chatting,
keep,
you know,
testing,
buy,
sell,
and everything
using our
Twitter account,
right?
right now,
you guys cannot
find Amy Swap
again,
but we
come back
stronger,
right?
right now,
we're launching
something called
ACAPers.
ACAPers
is actually
what we call,
you know,
more into the
rethink on
what we can
build
using the
stack that
we have
prepared for
the Amy Swap.
ACAPers
is not
just gonna
Twitter
swap,
port
Twitter
swap,
sorry.
My head is a
bit like,
it's just,
it won't be
just Twitter,
right?
But it will
be like,
even bigger,
right?
Because,
you know,
the past
problem with
Twitter,
right?
for the
first product
of ACAPers,
we will
focus more
on the
telegram
part,
right?
the fact,
which is,
you know,
it's actually
already live.
we plan
release it
real soon
in terms
of the
public
limited
beta,
right?
And what's
cool about
airport
telegram
is that
it's
a multi-chain,
right?
the next
iteration
would be
you can
buy and
sell
directly
from
group
chat.
it's
just,
you know,
you don't
need just,
you know,
chatting
via,
like,
you don't
need to
chat
directly
aika
swap
bot,
right?
But if
you bring
her to
your
group
chat,
right,
anyone
aika
swap
and buy
token
directly
from
group
chat
without,
you know,
even
need to
having
like,
individual
chat
or anything.
although
it's also,
you can also
buy from
directly.
But ultimately,
our goal,
right,
it's not just
building
another,
you know,
like,
telegram
20-week bot,
right?
we want
aika to be
like,
the ultimate
crypto app
that can
totally abstract
the whole
experience
between
users
and crypto,
right?
I assume,
like,
all of us
here are
crypto native,
right?
it's been,
like,
daily habit,
you know,
we know how
to use
Metamask,
we know
how to use
a screener,
we know
how to
check the
website,
smart contract,
and everything,
right?
But,
in reality,
right,
of the world
population
cannot do
what we do
on daily
basis,
you know,
they have
no clue
what is
chain,
you know,
what is
the difference
between
Arbitrum
and Solana
and Doge,
you know,
they have
no clue,
right?
And,
for the
first time
since I've
been in
crypto,
like,
in 2017,
right,
I feel
and I,
like,
I really,
I really,
like,
feel that,
you know,
by integrating
AI and crypto,
I think this is it.
This is,
like,
the ultimate
combination.
This is the
final abstraction
layer that
we would have
to bring crypto
into the mess,
right?
And I'm
very confident
within one,
two years,
right,
everyone,
you know,
like,
the majority
of people
that knows
and getting
in touch
with crypto
will be
coming towards
AI core product.
It could be us,
it could be
something else,
but it will be
something with
AI core,
you know,
that totally
obstructs the
whole experience
with chain,
website,
wallet,
and everything.
yeah,
that's
more,
yeah,
just,
we also will have,
like,
next Monday
if I can share
also,
right?
yeah,
just go into
my account,
look into
ECA first,
and yeah,
check our update
about our project.
I think it shows
a lot of
grit out of you
guys to
pivot after,
well,
after being
forced to
pivot,
frankly.
But,
I don't know,
correct me if I'm
wrong here,
Raz,
but it kind of
seems like
because
that initial
count got
banned,
you're able
to see,
like,
okay,
we had a
really cool
idea.
How do we
expand on this
and make it
much,
much bigger?
Yeah,
yeah,
that is true.
You know,
it's been two
months,
it's been only
two months for
but me and,
you know,
my team,
like,
feels like it's
six months
already,
or a year
already.
Like,
we are tired,
we are super
burned out,
and everything,
right?
But,
I mean,
at the end
of the day,
right,
like,
we kind of
see,
you know,
what would
become the
eventual result,
you know,
like,
the eventual
development of
AI is pretty
clear,
right?
Like,
everything is
going to be
AI-powered,
you know.
Previously,
we removed
the practicality
and now,
we will remove
complexity from
every single
thing.
It's not
just crypto.
Like,
AI will become
the biggest
bubble we've
ever seen
in our
lifetime,
probably,
hopefully,
right?
that makes
you know,
thinking,
like,
what we can
build,
what should
be our
goal,
you know,
not just
together,
right?
We don't want
just to
facilitate
a swap,
right?
We want
people to
basically
talk with
Aika on
daily basis.
We want
Aika to
skip all
the,
you know,
daily habit
of people,
you know,
which is,
you know,
like,
if you,
if you,
if you,
like,
play crypto,
right?
There's a couple
of daily habits
that you will
always do,
which is,
you know,
like,
first,
checking price,
checking your
portfolio,
checking your
shitcoin watch list,
right?
It's on the
screener,
define or anything,
right?
And then,
checking Twitter,
you know,
what are people
talking today,
you know,
what kind of
fight that I,
that I,
that I can see
today,
right?
And to be able
to participate
in all of this
activity,
right?
It requires hours,
right?
People spending,
like,
hours,
probably close
to 12,
18 hours
every day
just to,
you know,
keep in touch,
to track
every single
crypto development,
right?
And I think
the whole
complexity
can be just,
you know,
like,
you know,
how do you
call,
combine into
one,
right?
And then,
you know,
just compress it,
you know,
into a very
simple activity.
And that's
what we aim,
right?
other than
just the,
the swap
agent,
right?
We're also
building something
called ICAXPT.
So ICAXPT is,
well,
I mean,
I haven't seen
the ICAXPT
dashboard,
right?
but ICAXPT,
right?
It's,
it's,
it's more like
KOL dashboard,
right?
So it's a,
it's a place
where you can
actually see
the actual
every
call,
every shield
being mentioned
on Twitter,
right?
your favorite
influencer,
right?
You will be able
to tell them,
to tell whether,
you know,
they're dumping on you
or they're giving you
like a good,
good thing,
you know,
good,
good recommendation,
right?
So that's something
that we've built also,
right?
And,
that will also
relate with the
swap agent
since,
our end goal
of ICAXPT,
right,
is actually to,
give a recommendation.
So rather,
you know,
scrolling endlessly
through Twitter
for hours
and hours
and hours
and scrolling endlessly
to every single,
like,
tracker,
like,
wallet tracker
and everything,
right?
So we will just
get all of the
information from Twitter,
right?
Get it into a very
nice-looking dashboard
and automate it for you.
So you will get,
you know,
hey,
this user
that has,
like,
kind of,
like,
historical good performance,
right?
will actually tweet
something about this
and it's just not
just copy trading,
which is usually
just one user,
but you're copy trading
like a bunch of,
users.
And yeah,
it's still very early phase,
right?
But as of now,
we've been able to track
around,
like,
3,000,
data point
from,
3,000,
data from around
3,000 user
on this,
on this early stage.
And yeah,
it will be,
it will be fun.
It will be crazy.
Honestly,
it's,
it's the,
this whole,
AI crypto,
DeFi and everything,
right?
It's,
it's gonna be crazy guys.
I mean,
what we've seen right now
it's nothing compared to
what we will see
in like 12 months
or maybe six months
from now.
That's just,
that's what I believe.
That's my,
you know,
essentially,
yeah.
Sorry for the rambling.
No problem,
sir.
No problem,
sir.
We love having you on here.
We're really excited,
by the project
and super glad to have,
have you as a friend of,
friend of Arbitrum.
I really liked the idea
about,
about being able to,
you know,
trade,
trade directly from Twitter.
there's a lot of things
that I think
will be made a lot better
by AI agents
that can just carry out
actions
prompted
directly on Twitter.
And I don't know
if that's
powered by Grok,
if that has to go
through Elon.
obviously,
Elon took a little bit
of an offense
to Amy swap,
which is why
he banned it.
I mean,
I'm sure he,
it wasn't him personally,
but you get my point.
but yeah,
I'm excited for things
like being able
to use decentralized
Parlay protocol
directly from,
from Twitter
or being able
to swap directly
from Twitter
you know,
copying a trade
directly from Twitter.
It already exists
in Telegram.
why not bring it
onto,
bring it onto Twitter?
That's where you find
the tokens
in the first place
a lot of times
anyways.
But yeah,
thank you so much,
Raz.
I think next up
we were gonna have,
a Fed
from Flow.ai.
Where are they up?
I don't think
I see someone.
For you,
I don't know
how to pronounce it,
Fed.
Can you make me
a co-host,
Tony?
yeah,
I think the arbitrary
company needs
to make you a co-host.
I see.
Okay.
Okay.
Okay.
I see.
I see.
I see.
Okay.
I see what you're saying.
Hey,
what's going on?
How's it going?
Hey,
Fed.
First question,
how do you pronounce
your,
your username?
like the,
like the,
the federal,
the Fed.
just Fed.
I know how to pronounce
your,
your name.
Okay.
what about your,
your name on Twitter?
that one.
Oh shit.
I realized now,
that's the bronze age,
Ethereum.
So I put my,
my ENS.
Yeah.
I need to change that.
It's F E U L F.
it's my first name
and last name.
I don't know how you pronounce it.
I never pronounce it myself.
I see.
I see.
Gotcha.
Yeah.
I'll have to change it like back to the,
yeah.
How you guys are doing?
Doing great.
Doing great.
would love to,
let you tell the people a little bit,
a little bit about what you're building at flow.
Yeah.
Yeah.
Yeah,
of course.
And,
I'm happy also to talk about,
DeFi,
DeFi,
plus AI,
D F AI,
however you spell it.
But,
yeah.
So sharing about myself,
I'm an early engineer at gauntlet.
I was there for two years in the early days.
And I was also a team lead of the platform team.
And,
so full on,
focus on,
DeFi.
And then two years ago at Denver,
I start flow.
And the idea was to do a web three indexer,
but then,
I was using,
GPT 3.5 at the time.
So it kind of like people then evolve into these AI agents.
And today we are in DeFi marketplace or DeFi launch pad.
What I mean,
we build these agents,
that have different tools and they integrate fully with web three.
Specifically,
we build the AI agents,
that,
actually,
let me give you the,
the one that we're building now for Uniswap,
because that might simplify the conversation.
So for ETH Denver,
we're preparing these AI agent that is an ecosystem explorer for Uniswap and,
also a DeFi,
the F AI agent,
with intents.
And so you can use this agent to ask questions like,
like a swap one ETH for,
USDC or,
well,
these days probably you want to give the opposite example since ETH is like,
right now it's cheating the bad.
So maybe it's not a good example to change it for USDC,
but you get a point.
And,
yeah,
I'm very bullish on the DeFi space.
is there a way for me to share some blog posts?
I've been writing,
I wrote two blog posts on DeFi and I have a thesis that I think would be worth to share with the channel.
Yeah.
Send it,
send it to me in telegram and I can post it up here.
Okay,
great.
you're here.
Yeah.
I think you might even be able to pin it yourself as a speaker.
If you click on the top three buttons of the post.
Yeah.
I need to switch to my laptop.
Hopefully it won't make a mess.
Yeah.
Or you can just send it to one of us.
here it is.
We'll pin it.
Found it.
Found it.
Found it.
Found it.
all right.
I added it up there.
Yeah.
Okay.
that's the number one.
Let me send you the number two.
Number one is for a technical audience.
So if anyone in the,
in the listener is an engineer,
we'll find that interesting.
And my thesis is that,
the internet,
used to be deterministic or is deterministic.
And the reason why it's been successful is because it's,
it's a state machine to some degree.
Blockchain is a full state machine.
So everything happened in transaction.
And,
it's because of that is reproduced,
reproducible,
reproducible,
whatever you,
however you say the word and,
composable.
And so the reason why DeFi is composable is because it's a state machine.
And because of that,
we have MEV and we have flash loans and other characteristics that are unique to,
DeFi.
We don't have that in Trudify.
Okay.
it's more like moody and non-deterministic.
And you don't want to add a non-deterministic entity to like run something,
that like on top of an API,
because you want the API to be deterministic,
but,
you want to use the AI agents to do anything.
There is,
has to do with reasoning,
anything that has to do with sentiment analysis,
anything that has to do with,
any kind of decision making that as human are slow to make and can be done
immediately.
I'll give you an example,
like a Trump token that came out,
two weeks ago,
or a month ago,
I think now,
you could have a DeFi agent,
you give 1% of risk profile,
and,
you say this 1%,
you can use it for whatever,
what you think could be like,
a risky position or like YOLO,
whatever,
this is your YOLO money.
And,
you have this token coming up,
the agent could immediately do like a transaction and jump on like the
Trump coin.
so that's,
that's one example.
But,
the second article that I shared,
I actually have these other thesis that is,
what is the path of AI agents?
And where are we today?
And where are we going?
and,
for the sake of the room,
I'm sure there's other people that want to talk as well.
So I'll try to keep it a little bit shorter,
but the concept is the reason why,
chat GPD cursor and other LLM have been so successful is because they have a
feedback loop that is really short.
You ask a question,
you get an answer.
You ask her to change the code.
the LLM change the code immediately.
You can review it.
Great.
cursor,
for example,
the app like Devin that you just write the code,
the agent Devin that you write the code,
you ask to change something in your repo,
you launch it,
comes back 10 minutes later,
change your PR and that's a lot of stuff.
it's,
it's not successful today because the feedback loop is too large and,
the agent still has a lot of allusionation.
So there's a lack of like trust and there's a lack of like,
quality in the responses,
but over time,
we'll see like from,
today from the,
is this agent that you ask a question and get the feedback immediately,
we'll start seeing AI agents that have a larger feedback loop.
And so in the DeFi space,
how these,
how works in the DeFi space,
I think today,
what we'll see in the next month or two will be the AI wallets,
and so anything that is like intense,
swap me this coin,
send this payment to my team.
And I have ETH on Polygon.
They want to receive one in Arbitrum,
one in Polygon,
one in Linea.
And so the agent does all this transaction for you,
gives you back,
like a strategy and you sign it with MetaMask,
send it,
so far so good,
right?
So the next step I think will be the DeFi agents where,
and by the way,
if you open the article,
you can see the,
I design like a map where you can see these like steps.
The next step will be the DeFi agent,
which I think would be these same autonomous agent that runs on chain and do portfolio rebalancing.
They might even do things like risk management and open proposals,
but the human is still on the loop at the end.
So you still need to sign,
the final transaction.
And then I think the ultimate stage will be,
we all agree is like,
fully autonomous agents.
I have a twist on that and I have a very personal opinion on that on how we should do it.
I think those agents,
should have,
a governance system that uses something on the line of work on where we have a proof of humanity for the shareholders.
why that you don't want,
the,
the governance of the Dow to be with a share.
So like,
for example,
Aave,
an AI agent could all have it,
the,
the Aave governance token and could take over and just use it and do whatever they want.
Now,
if they take over Aave,
no big deal.
Aave keeps,
keeps running good.
But if an AI agent takes over itself,
I don't know.
It's like,
it becomes kind of like these,
situation in which like it becomes unstoppable and it becomes like,
autonomous,
fully autonomous and we lose complete control on it.
some people might be okay with that.
I think we should keep the human in the center,
but that's my,
my take.
And I'd be curious to hear what you guys think about this as well.
Chase or anyone else?
is that working?
Sorry.
I couldn't unmute for some reason.
yes.
thanks for the,
thanks for that fed.
The,
I've gotten a chance to see and play around with flow AI a little bit.
And one of the cool things is you can make swaps directly from the interface
while also chatting to an AI about,
Hey,
should I be doing this swap at all?
Which could turn into like an endless,
endless feedback loop of,
Oh my God,
should I make the swap?
Should I not?
Should I sell?
Should I buy?
I could see people getting,
getting caught in like a vicious cycle of questioning themselves,
but Hey,
that's fun.
what was your question?
yeah.
well,
I guess,
in this path towards AI becoming like fully autonomous,
if the vision,
the thesis that I have that eventually we'll have this fully autonomous agent
and we should keep the key for the agent.
And the way we can keep the key is having this war coin or whatever other token
that have like a proof of you of humanity.
So I wonder if that's something that resonates with you as well,
or you think it's okay for the agent to be completely autonomous and there's
not,
they can hold their own token if they want.
I not only think it's okay.
I think it's important.
I might plug my own blog post that I did the other day,
but I did a blog post on the lit protocol agent wallet recently,
which uses,
T E's and threshold in PC to ensure that no agent developer can rug the tokens of the agent.
I'm not saying every agent will fit this use case and every agent should have its own tokens
that humans don't have control over.
But if you're going to launch a token,
like,
I'll use a re bro as,
as an example,
and you're going to give the agent tokens,
you need to ensure that there's no trust involved.
It needs to be a trustless solution that the agent won't go or the agent developer won't
go and just transfer all of the tokens in that agent wallet to himself or to his Binance
account and dump the whole thing.
So yeah,
yeah,
that's an interesting point.
I know the interesting thing about lit protocol is it allows you to parameterize actions as well.
So you,
you,
you can say to the agent,
Hey,
only do this,
only interact with these tokens.
You can only buy this amount on any given day.
I guess,
I guess,
okay.
I guess the,
it's the kill switch,
basically what I'm talking about.
And I wonder exactly what that even mean,
but the fact that an agent could be completely autonomous,
for some reason I'm still not comfortable with that.
And it might be,
yeah.
And I forgot,
I think it was,
who was,
Yann LeCun or someone was talking about that AI will need to have a soul and we'll need
to just leave it free.
because if it doesn't,
it will do anything to be,
free.
And,
it might mean attacking us.
So yeah,
maybe the right way is to completely remove the key.
Even the kill switch.
Sorry,
but I just want to ask what,
I mean,
something that I tried to ask myself,
right.
But why does an agent need to be fully autonomous?
Why does,
or why doesn't?
Yeah.
Why does,
I mean,
what is it for?
What is the purpose?
Like,
yeah,
something that I,
you know,
try to ask myself,
because most of the AI agent is just,
I mean,
the,
the main purpose of the AI right now,
right.
This extension of human since,
to be able to run an AI agent or any AI related stuff,
right.
someone needs to pay for it.
Need to cover the cost,
right.
Cause,
right.
And ain't nobody gonna like build their own AI,
right.
Right now people are using like assisting alarm for it,
no matter what it is,
right.
Essentially someone needs to take care of it.
It's to feed it and yeah,
sure that,
you know,
the credit card is being paid to subscribe.
So I like that brings the question,
like why,
why,
why does an AI agent need to be autonomous?
Like,
and it's possible for AI agent to be fully autonomous or be just pretending that the agent is autonomous.
That's something that I,
I try to ask myself,
well,
well,
you know,
like building this,
like,
yeah,
cause that's something that people keep talking about,
like the AI,
you know,
AI agent that can do self trade,
a swarm agent that can launch token and,
right.
getting,
income on their own.
Like it's honestly like,
it sounds to me just,
it's not realistic.
As of now,
I don't know.
Actually,
interesting.
So that's why I'd like,
maybe that's the discussion.
Yeah.
I'd like to discuss with you about this one.
yeah,
I think actually it's not only possible,
but kind of inevitable.
I think that where we are today,
it's like,
we just saw like,
I think it was John Carmack was talking about where six,
breakthrough away from a GI.
And now today we're probably at four or three already,
like a deep seeker one with the reinforcement learning,
unlock the age,
the LLM to learn autonomously.
So you just give,
reinforcement learning,
model.
You give rule,
a rule.
The rule is you need to solve this math problem,
for example,
and the,
and the AI keeps like a testing one plus one,
two.
Okay.
Now two plus two,
four.
Okay.
And it keeps going and now makes mistakes.
When makes mistakes,
move on and,
uses,
Monte Carlo tree search.
So whatever path in the decision making is wrong,
it goes back and goes in the other one.
And you end up with like models.
They are incredibly smart.
they're all one level,
like deep seek.
R1 is,
open AI or one level in math.
It's not in other benchmark like the Alice in Wonderland problem,
but is in,
MMLU and both in also in math and other,
like,
more scientific,
like,
topics.
And so what's going to happen?
I think the agent today,
we need to see them.
We need to give them,
like,
training sets.
We give,
we give them content,
but eventually they will become like learning on their own,
on any kind of rule.
That we can give them.
And the other reason why I think the evolution is like inevitable is,
all eyes are on it or money are on it.
And,
it's a compound evolution that is happening both on the hardware level,
on the software level,
research level,
economic level.
Crypto is pushing,
is pushing on it as well.
So it's,
everyone is pushing on this.
So what,
why,
to answer your question,
why is,
is necessary to be autonomous?
And if it's in,
so if it's possible,
yes,
it's possible,
but why should be autonomous?
I mean,
you can see like,
the Kyle,
some money,
some money pieces of,
zero employee companies.
I don't know if you guys read about that one.
basically he's saying that,
and,
and we see that in tech,
the trend is company are shrinking.
You don't need a team of hundred engineers,
like 10 years ago to run,
servers because now you have Kubernetes.
So now you need 10.
And then now Kubernetes can run with an agent.
So now you need just one or two.
and it keeps going and it gets like a shrink.
The,
the size of a team tech team,
especially shrinks and it's going to zero.
And,
the moment it goes to zero.
Yeah.
The agent should be autonomous where,
you basically give directions and the agents does whatever need to do that.
It's,
under your guidance.
And at some point it would be even completely autonomous.
For example,
another example could be a portfolio manager.
You don't want to manage your own portfolio,
every day,
maybe sometime you want,
but if there's like a,
an autonomous model that just run continuously,
why not?
Yeah,
but it's,
I mean,
it's just like,
it's been,
I've been in crypto for,
for maybe too long,
right?
So anytime sometimes,
something's a spectacular coming out,
it always,
I always trying to be,
you know,
like,
is it too good to be true?
Cause mostly everything in crypto that is too good to be true.
Usually it's too good to be true.
Right.
So I,
so I think,
I think there's a visual,
eventual end goal,
like what you mentioned,
right?
But the gap between,
what exists now and the eventual,
the inevitable,
ending or inevitable,
like outcome,
final outcome,
right?
It's something that I,
I don't think we,
we know,
you know,
like the gap could be two years,
it'd be 10 years,
20 years.
Oh yeah.
Yeah.
Like in a couple months.
Yeah.
Absolutely.
So the,
the time,
yes.
The time we tend to overestimate the,
our prediction tend to overestimate the short term and underestimate the long term.
So in the short term,
yes,
we're not going to see these crazy agents yet.
Maybe even two,
three years could be,
but in the long term,
it's,
yeah,
it's going to happen.
And,
by the way,
I'm in crypto through like,
as well through like working in different,
web three companies and running the Bitcoin NYC community.
but I'm also in AI through these AI NYC community.
And,
I've been around,
a lot of other experts in AI that are way smarter than me on,
from Google,
open AI,
meta.
And yeah,
it's the AI evolution right now.
It's not really on the,
on the crypto field is really in every single field you can think of.
There's not a company that doesn't try at least to use the buzzword.
Yeah.
And it's a buzzword,
but if you remove the buzzword,
this is one of those cases where,
we have a real use case and it's tangible.
It's not even like,
I don't know.
yeah,
it's not even just a buzzword,
the buzzword,
the DeFi,
that's a buzzword,
but the,
the concept of what it will be in the future makes sense.
What will be today?
today agents are shit queen wrappers or wrappers,
LLM wrappers that,
have a shit coin,
on a Solana,
pump fun and they need to pump it.
That's what it is.
And,
if they,
they do jack shit,
they just tweet random things and,
probably,
token terminal and other ones,
they don't even tweet.
They just have someone that tell them what to say.
I mean,
it's,
let's be honest.
That's,
of course,
it's like,
those are narratives.
but,
the,
the application are real and we're going to see like a big,
yeah,
we're going to see a big wave of,
real application.
And finally consumer apps in crypto that we've been talking for the last
few years,
where we've been talking about infra,
infra,
infra,
we need the applications.
and,
just a side note.
I think that Ethereum need to capture the DeFi space or it's completely
fucked because,
the meme coin space and attention space is gone.
Solana got that one.
The tech space Solana also is pretty good as a platform.
And then,
tokenization BlackRock is about to tokenize the shit out of the,
of the real estate and,
and probably also the stock market and,
cut on Bitcoin is about to become the tokenization platform.
when Bitcoin adds,
the BRC 20 for tokenization,
and,
they start implementing things there.
Why should you use Ethereum if Bitcoin is more,
more stable.
And so,
I have the,
that's my opinion.
And I'd be curious to hear the room,
but I think that Ethereum need two things.
One,
we need to capture the DeFi AI space.
And two,
we need Vitalik to go to a fucking gym,
jacked up and going,
he's,
like a Brazilian Jiu-Jitsu arc and become like,
like,
Zuckerberg.
I'm joking on the last one,
but Chase,
Chase is already jacked,
bro.
Chase,
you're already jacked.
And what I was going to add to that,
bro,
it was like,
our RB,
we're going to do,
we're going to do,
DeFi AI on RB big time.
That's,
that's what our bedroom is now,
let's go.
Yeah.
And that's,
that's kind of what,
you know,
I was listening to you,
to you talk about and,
you know,
big fan of what's going on in the flow.
also,
also based in New York city.
So if there was a time that maybe we can connect,
we could DM after this,
but like,
I think that arc of AI driven instruments,
you know,
on the Ethereum blockchain,
it's definitely incredibly relevant.
And there's a lot that can be done and should be done,
especially as we like make this ascension towards cross chain
interoperability and make it seamless across Ethereum.
one idea that,
you know,
we were talking about earlier,
I'm not sure if you were,
you were on,
but one of the ideas I was talking about with chunk was developing
some type of AI powered cross chain yield optimizer,
especially considering that the fragmentation between yields
across multiple L2s will decrease as interoperability expands.
I think that there's like a really interesting idea where you can
have an AI trained system.
That's essentially aggregating or showing like where the best yields
across chains are at all times,
where the user accesses a vault,
those integrations are already done within that vault.
So you have X amount of pools that are already integrated.
And then the AI is just showing you which five or which six that you
could allocate to in real time or automatically like just,
just yielding you assets.
Yeah.
And it's like,
it's an idea that like we just kind of started exploring through,
you know,
different index features and it's trying to take like yield bearing
index features and taking them like to the next level.
Like,
okay,
like,
yes,
we can tokenize a basket of yield.
That's,
that's not like,
it's,
it's,
it's very helpful.
It's very cool,
but that's not taking it to like the next level.
How do we take that to the next level?
So if after this,
if you have some time I could DM you,
maybe we can connect and share a little bit about the idea and happy to
meet up and talk about it.
Yeah,
please.
I don't even know who you are,
where you are.
Cause in the,
I don't see the speaker who's speaking.
I'm,
yeah,
my bad dude.
It's a,
it's Joe with a cryptex finance.
Okay.
So I'll send you,
I'll send you a DM.
Okay.
Awesome.
In the,
in the fuck here in a spaces,
I can see like where the voice is coming from.
Yeah.
so I'm a boomer.
what was I saying?
yeah,
it's pretty much is what we're building with the Uniswap.
It totally makes sense.
And the idea is exactly that you're going to use something like a volts.
That F F Y I,
which is like,
yeah,
you have all this yield and then,
you pretty much,
you can explore the yield with the agent as,
which is the best one for your risk and risk profile.
And based on that,
the agent will do a multi-step transaction,
from your wallet or go from a ramp,
then a swap,
then bridge,
do all the things and then go.
Yeah.
So absolutely.
That's,
totally like,
where we're going.
And,
it it's,
yeah,
it's a combination of like using the data and using the transaction on
chain.
Now,
the question that I have for you is where the,
where is the mode?
Cause we're building this and I'm sure other probably hundreds,
hundreds,
founders are doing the same and they might do it on a different,
slightly different,
angle,
but I mean,
everyone is building this shit.
So how,
what is the differentiation?
What is the,
the mode?
And I've been squeezing these,
question for the last two years.
And I think this one will be useful.
I think for the room as well.
and I think BC as well,
kind of like,
agree on this one.
Distribution is number one,
obviously.
If you have distribution,
you're Metamask,
you're a wallet,
you're Coinbase,
you're sitting next to the customer.
that's,
your biggest mode.
And the Google right now is not the best one in AI,
but they have the biggest mode because they have the biggest distribution.
Okay.
And then,
that's one.
The other one is data and data is going to be essential is the oil and it's
going to be essential.
But now with tools like fire crawl or,
even deep research,
you can pull any data whenever you want,
just,
one prompt away.
So you do ask,
give me this data and put this in my database done.
And you have it,
we build that,
but,
three months,
everyone will have that.
So even the data,
if it's public,
you can have the data.
So there is a mode if it's private,
if it's public is not.
And then the other modes will be private models.
and I think in real life community might be a big one,
but then in real life community is a mode.
there is a little bit,
I don't,
I don't want to say weak,
but,
how big can be a community and how profitable can be.
yeah,
that's the biggest question that I have right now.
What is the mode and how do we,
make this thing survive?
absolutely.
And I think,
I think to your point,
you know,
one of the things that I just heard you say was with regards to
distribution,
you know,
of the asset,
obviously,
you know,
the bigger you are,
the more easily you're able to distribute,
you know,
what you create.
But I think,
you know,
all of us here understand that a lot of the concepts that we
discuss collectively as builders are far beyond the normie or
the regular users comprehension in terms of,
in terms of what we can create.
And I think the teams ideally that do succeed in the integrations or
the,
or the concepts that we're discussing here today are,
more so the teams that can break that down,
simplify it and just make it a seamless experience where you don't
even know any of this shit is going on.
You just access a UI,
you access a system,
you interact with that system and the system does what it's set to
do by the defined parameters of,
of said system.
And I think whoever can make that approach as streamlined and as
easy as possible,
are they guaranteed to win?
but will they have some relevance?
Absolutely.
Especially if there's a community behind it that is supportive of,
of what they're working on,
then they can try to scale that into,
into bigger and bigger ecosystems.
But to your point,
the key takeaway that I got from,
from what we were just,
what I was just listening to was like distribution,
1000% incredibly important,
how you structure the UI and the interface to be super intuitive,
you know,
especially behind the scenes without the user really even knowing,
you know,
what's going on,
but,
but more so like scaling it and,
and,
and doing different things like within the networks that just,
just make it like super easy to use,
man.
So ideally,
ideally that,
that would kind of be like how,
how we would approach that.
And,
you know,
understand like,
there's always going to be a thousand different teams,
you know,
building a thousand different things.
It's,
it's great for the ecosystem as a whole,
but we all know out of the thousands that are building,
not,
not everybody's going to,
going to succeed at it.
And,
you know,
there's,
it's just,
it's,
it's going to be like that,
whether,
you know,
we're,
we're building blockchains or we're baking bread,
you know,
there's,
there's going to be people who gravitate towards,
towards certain things.
And,
I think collectively as an ecosystem,
we win,
maybe not as individuals so much,
but as an ecosystem by pushing these ideas forward,
and just trying to capture users and bring them within,
within our communities.
And man,
you know,
like just got to be honest,
like really impressed with the things that I've been hearing from you on
this call.
You sound like,
you know,
you're really sharp in,
in your knowledge base and the things that,
you know,
it was very informative for me.
And,
really thank,
Like happy,
happy to listen,
really,
really great stuff from you.
Honor to,
to hear that.
yeah,
just,
just thought to,
what you were mentioning about,
to like elaborate on what you were saying about building the UIs that,
are seamless.
yeah,
the,
the reason why onboarding has been so shit in crypto is like,
to do anything though,
you need metamask to sign,
any single steps you need to install 10 different things,
get token from coin base,
then send,
I mean,
this is the kind of stuff that our,
boomer family kind of do,
let's be honest.
And,
and I think,
1000%.
Yeah.
I know the reason why the,
the Trump coin work is because you could just get the Solana wallet,
put some money,
buy,
done,
pump the fund,
whatever.
so the question is what will be that in,
in Ethereum.
And I think there's a,
well,
the,
I need to look more into these,
but,
abstraction,
wallet abstraction.
That's one that could be helping.
And then there's this ERC called seven,
seven,
seven one five.
That,
pretty much you give a low answers with policy.
So you can say to a smart contractor,
you have $1,000 a month and you can do whatever you want.
So you don't have to sign every single transaction.
You just send it once and it's free.
And I think we need something like that,
where you just use the credit card,
put the coin,
whatever on your wallet.
And then you don't even know what you have there.
You just write some phrases or talk and it just goes.
yeah,
I think,
I think directionally you're right.
We'll have to get there.
But the question also for you guys is,
do we focus on the onboarding of the new users first or on the power,
power users?
And,
and that's the other question,
like the other million dollar question.
I'd,
I'd like to actually add to that,
this is a Buccio from chain craft.
Cause that's something that we always think about is the sort of friction for new people to the space.
so just real quick,
what we're doing at chain craft is,
basically building better tools to create games and monetize them easily.
So we're kind of doing something that isn't is built on the blockchain,
but is kind of our long-term audiences is not blockchain native.
And one of the things that I feel hasn't been used enough that it's just such a simple thing,
but can really stop a project in its tracks is like fees,
like even just the gas fee.
And,
I just want to throw that out there.
Cause I don't think I hear that enough,
especially as like gas is becoming cheaper in some of these like L3s,
like Senko,
like a few rats here,
a smooth phone.
I feel like more people should consider that as a way to also just simplify,
just starting to do transactions or events that are done on chain.
and we definitely going to be implementing that on our project.
That's amazing.
Sweet guys.
I just want to make sure we're,
you know,
we're on time here since I think we're like 15 minutes over,
you know,
sketch.
I appreciate everyone here for coming on,
talking about their projects,
especially,
you know,
that last little conversation there.
I think ideally these spaces are,
you know,
it's,
it's time to like,
yeah,
introduce what you're building collab potentially,
and even just kind of talk about some of these existential questions,
but not getting too deep.
cause I mean,
frankly,
you know,
like we,
we all work in crypto and,
and I,
you know,
adjacently AI,
cause we all believe in,
you know,
ideally a decentralized and permissionless future.
you know,
we're all building towards that,
but definitely the idea here is we're,
we're building on Arbitrum,
DeFi,
the intersection,
both,
of course,
appreciate everyone here for joining again.
If you want to connect with anyone you spoke with,
or anyone you heard for anyone in the audience,
feel free to DM them directly.
DM me if you want chunk,
happy to help you guys connect,
after the fact and make sure that the,
the ball is still rolling even after the call ends.
Yeah,
guys,
thank you so much for jumping on.
Appreciate it.
And chunk,
I'll connect with you in a bit,
man.
