Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
Chainlink Runtime Environment: Redefining Interoperability ⛓️ The Aggregated Ep. 144
Recorded: Jan. 9, 2026 Duration: 2:05:36Space Recording
Short Summary
In a recent discussion, industry leaders explored the launch of innovative projects and partnerships in the DeFi space, highlighting the integration of real-world assets and the emergence of risk-managed pricing strategies. Key players like Llama Risk and Judge Finance are collaborating with Chainlink to enhance yield opportunities and improve risk management in decentralized finance.
Full Transcription
Thank you. Music Thank you. Music Thank you. Thank you. Music Thank you. Thank you. Thank you. Hello, hello hello hello
and mic check mic check can everybody hear me nope no wonderful loud and clear loud and clear. Loud and clear. Wonderful, wonderful. Welcome. Welcome, everybody, to this week's episode of The Aggregated with our esteemed guest, Chainlink.
Yeah, welcome to the show here.
As you all know, here at The Aggregated, we like to kick it off here with some intros here in the beginning.
aggregated, we like to kick it off here with some intros here in the beginning. Today's show is
going to be about building across chains and systems with our Frontier Chainlink,
and this is the CRE portion of it. So if everybody can go ahead and introduce themselves,
I'll just go left to right here on my screen
of what I'm looking at, and we'll start off with Patrick.
Hi, and thank you for having me. I'm Patrick. I'm the CEO of Ampli Technologies,
and we build validation system for agentic capital management.
Wonderful. Do you want to tell us a little bit about agentic capital management?
Sure.
Like if you pay attention over the last year in DeFi,
you might have noticed this emerging space of projects
offering agentic intelligence in order to optimize
across DeFi lending markets.
And this is where me and my co-founder Xavi, we got very involved early as we were advising
Giza, that's one of the main players in that field.
And we were actually starting to scheme how we could use these types of agents in the
back end of consumer applications in order to abstract
away all the crypto complexity.
And while doing that, we realized, wow, there's actually a whole middle layer missing here
that needs to be built in order to make this ready for institutions, but also these huge,
large scale consumer applications that make sure that we can actually trust the agents
with the capital that we put
into into their hands basically a system of checks and balances and that's what we started to build
that was around like a year ago and yeah now we're we're ready to ship that wonderful thank
you for that and welcome to the show here uh next we're gonna we're gonna go on over to wormhole hi thanks for having me i'm a wormhole
founder of llama risk so llama risk is a defy risk manager and we work with some of the biggest
names in defy so some of our clients include ave ave horizon curve finance and athena and the work include Aave, Aave Horizon, Curve Finance, and Athena.
And the work that we do is generally
to advise these protocols on how they can have
safe and sustainable practices so that they can grow sustainably
for the long term.
We do things like help them with parameterization
of the protocol, asset onboardings.
And more recently, we've been working very closely with the Chainlink Labs team to develop solutions on Chainlink CRE.
We're working to develop solutions around risk-managed pricing and automated emergency response that can be integrated into DeFi protocols,
including some of the partners I mentioned, but for DeFi generally. So thanks for having me.
Yeah, wonderful. Thanks for joining. Next, we're going to go on over to Swapper Finance.
Hi, it's Christopher here. Thank you very much for having me here as well. Swapper Finance is a next-generation payment and onboarding infrastructure layer we built in collaboration with MasterCard and Chainlink. direct user deposits from card payments, web-free wallets, or crypto transfers from centralized
exchanges. The goal is to propose one simplified, single- and compliant, end-to-end solution
that helps protocols onboard their users in multiple ways, from fiat through crypto,
directly into their DeFi applications. Wonderful. Thank you. Thank you and welcome.
Next, we're going to go on over to Austin.
Hey, thanks. My name is Austin Kribbysh.
I'm a co-founder and CTO at Instructs.
And we are focused on on-ramping the next wave of real-world assets into DeFi.
And our most recent work has been around the precious metal USD
stable coin that you've probably seen recently in the curve ecosystem.
So a lot of the technology we built around the chain
is helping drive the underlying gold that's backing that stable coin.
Good to meet everyone.
Wonderful. Well, welcome.
Next, we're going to go on over to Charles.
Hi, thank you very much for having me.
I do appreciate it.
So my name is Charles Holmes.
I'm the co-founder and CEO at Judge Finance.
We're currently building a yield layer for stable coins, targeting low-risk DeFi.
We're Chainlink native cross-chain operating
system powered by CRE and CCIP and Consero for secure automated optimization of yield.
Our core product, Yieldcoin, is a permissionless vault that takes stables such as USDC, USDT,
from any chain and then reallocates them to the safest, highest risk adjusted
opportunities across any protocols.
And we stream the net yield to users after a transparent fee.
Yeah.
And so I really appreciate you for having us today.
So thank you very much.
Wonderful.
Yeah.
Thanks for joining.
Zen, is it Zen?
Are you there? Zen, is it Zen?
Are you there?
Is it me or can anybody else hear him?
No, no, you're good. Yeah, I can see that he's unmuted, but yeah.
Okay, well, sometimes we have a little bit of difficulties with X,
but hopefully we can get that fixed. You might want to drop down Zen, maybe pop back up.
I can maybe quickly introduce him. He is Ampli's CTO, so he's my colleague, and I hope he has his audio issues fixed in a minute.
Okay, wonderful. I see Ash popped up here. Ash, do you want to go ahead and introduce yourself?
popped up here ash do you want to go ahead and introduce yourself yeah for sure um great great
for thanks for putting this together i'm happy to be here so i'm ash i'm part of the strategic
initiative team at chain link um right now focused on chain link runtime and chain link automated
compliance engine which is ace which is a product that is built on top of the chain link runtime
and as um you guys will see with this new environment,
we are seeing very cool new use cases coming alive.
So yeah, very excited to be here.
Wonderful. Welcome, welcome.
And whoever's behind the Chainlink account,
do you want to introduce yourself?
And maybe, I know everybody knows about Chainlink.
I don't know if anybody in the space,
if you've been here for any amount of time,
you would know who Chainlink is, but
do you want to introduce yourself?
I think that account might just
be Aura Farman.
I see.
Okay.
Alright. Looks like
Zen's trying to pop back up here really
quick.
Are you with us now, Zen? Yeah, I'm with you. I'm with you. Can you guys hear me?
Yes, we can hear you loud and clear. Welcome to the show.
Yeah, great to be here. Yeah, so I think Patrick did intro me, but you're working very closely
with the CRE team and with CRE. Specifically under what we've done is we've been able to abstract
the centralization requirement of agent execution.
So that's been really, really fun to work with
and really excited what we're building.
Wonderful.
And my name is Richard.
I'm the Chief of Staff for Lunar Digital Assets,
which is the venture city known for incubating unicorns
like Polygon, QuickSwapap and various others as well.
I want to give a shout out to Rock down there.
He's the CEO of LDA and he's also the co-founder of Quickswap.
I see him down in the crowd.
It's probably really early in the morning for him.
morning for him. And then Darren, do you want to introduce yourself?
And then, Darren, do you want to introduce yourself?
Yeah, Darren here, head of ecosystem at Quickswap, most commonly known for masquerading as a panda.
Yeah, obviously Quickswap. I don't think we've formally announced it yet so it might be alpha but um i think either right now we're ready
or we go live on monday we'll uh we'll be going live with with chain link cgip to get quick from
polygon to base to ethereum uh we'll start uh that process have it on quicksop pretty soon
wonderful that's that's big news that's major alpha, at least for me, for sure.
I want to welcome everybody here to the show. If you guys haven't been on the aggregated before,
I'd like you to think about this as more of like a Roman bathhouse style.
When I go to throw in a question here, feel free to jump in. No need to raise your hand or anything.
Just let the thoughts just roll off the top of your head, and that way we can get the highest
level of conversation. So the very first question that I have for everybody here,
just to kick this off, when we talk about building across chains and off-chain systems, what does that actually mean from the perspective of real productive teams?
I don't know if anybody can just jump in there.
Yeah, please go ahead.
Yeah, I'll kick this off. Okay, so for Judge, it basically means creating resilient, scalable infrastructure
that abstracts away the fragmentation that we all see in blockchain ecosystems
and at the same time, integrating off-chain computations for tasks which are impossible,
inefficient, costly, these types of things.
Further, it means all of these contracts that are deployed on these various chains
can come together and make a cohesive system um it means that chains which have specific
constraints they're abstracted away and with cre we really think that all chains and all off-chain
components are now basically abstractions so with our yield coin system we're going to take that a
bit further and so we use that to ingest stable coins from any chain.
I use off chain data such as APY data, historical data,
liquidity data from DeFi protocols.
And we use that to compute optimal allocations
and executions across chains, protocols, and tokens.
So yeah, that's what that means to us,
just to kick that off.
If I may jump in quickly. That's a very, very good question.
I think it's the off-chain plus on-chain connectivity,
it is something that was always the case, right?
Being able to connect data from off-chain
with the data that is happening on-chain, allow transferring data and user across those layers was always the case in crypto.
And right now with the Chainlink runtime environment, we do have a right tooling that can
help executing it, help building those workflows that can in a decentralized manner
executing it, building those workflows that can in a decentralized manner
provide us the way of connecting data from off-chain sources with the data from on-chain sources and
executing transactions or the workflows itself, right? And this is exactly it for us at Swapper
Finance. The utilizing theory here is about orchestrating is about
getting to know when something happens and that actually off-chain I get some off-chain routing
for the best best path for the user things that you cannot do on train in effectively but you can do off-chain and utilizing CRE workflows, you can do it very
efficiently. And that's how I see it. And that's how I believe it impacts this production
things.
Cool. I'll just throw in from the RWA perspective. One of the interesting things about CRE is that we're able to,
well, first of all, RWA is a very compliance and data hungry process. So for us, CRE is really
important to connect those off-chain systems with the compliance logic that's required by our
customers in order to drive token issuance on chain. So from a production team perspective,
things like middleware like Chainlink is highly important.
And quite honestly, I'm not sure if there's a future for RWAs
without the off-chain processing that Chainlink CRD is capable of doing.
And from LumaRisk's perspective,
we're very interested in the trend with RWA starting to come on
chain more.
And more generally, we could categorize anything that has off-chain dependencies or off-chain
reserves as being an RWA, which includes all of the stable coins, which has been one of
has been one of the big narrative drivers of this past year.
the big narrative drivers of this past year.
So from our perspective, with more and more of these real world assets coming on chain,
the importance is to maintain transparency standards and guardrails for safe integration
of these assets into DeFi.
So we're always working on the intersection of DeFi and TradFi as asset issuers are looking to come on chain and's a very robust way of deriving safe pricing for those assets on chain. regulated assets that may have white listing requirements, this is particularly important
where you may not have a diverse set of sources where you can derive a fair market value for
those assets and you may rely entirely on the asset issuer themselves or the fund administrator
who's doing the accounting for the NAV of the fund.
So these are the types of questions that we're looking to resolve as we're seeing a greater
and greater reliance on off-chain sources integrating into an on-chain environment.
Yeah, I think that's a great call out from an RWA perspective.
The triangulation of data compliance and risk is kind of the sum of all of those.
And Chainlink is your stablecoin issuance.
So cool call out.
Just to quickly round that off from our perspective, since we are mostly concerned with intents that live off-chain
and are generated off-chain, but then influence
on-chain transactions, our job is
to manage this centralized point of failure, potentially,
while avoiding introducing new centralized points of failure.
So that's oftentimes one of the big challenges when we're
going cross-chain or from the chain
into the real world and back is those
are the most dangerous attack factors
that we have in our industry.
So our job is to make that as secure as possible.
All right.
Does anybody else have any input on this one?
Okay, we'll move on to the next one.
So interoperability is often framed as simply moving tokens between chains.
But each of you is solving something deeper than that.
What are the hardest hidden problems, data consistency, roles, permissions, risk validation, orchestration, compliance that teams often underestimate?
Once again, guys, just feel free to just jump on in.
I can jump in if you can hear me.
Yeah, I can hear you.
Beautiful.
So another good question here at Swapper Finance,
although we are not only focusing on interoperability
in the sense of on-chain interoperability,
we also enable different ways of depositing for our users including
card payments fiat onboarding etc uh strictly about inter interoperability i think the biggest
challenge is about handling multiple use cases or edge cases what will happen when something goes
wrong how you will get the best quote for the user if you are not only transferring,
do simply bridging of the funds, right?
If you want to execute smart contracts through programmable token transfers
via chain link CCIP, how you ensure that the message when it will arrive,
the destination blockchain will still be a valid one.
Those challenges are usually not seen
from the user perspective.
The end user wants to have a nice and easy flow
from the user experience perspective.
And it is on us as builders to solve those challenges
to ensure we would have
a proper solution for that. And here CRE definitely can help you,
especially with the orchestration chart,
especially with getting, reacting
on what is happening on-chain,
and fetching additional data of chain,
and then again, interacting on chain.
This would be, I believe the biggest challenges
we have seen from our side,
but I'm super happy and curious to hear
what were the challenges other teams were facing
or are facing.
Yeah, so from Judge, our biggest challenge at the moment was moving between assets.
So everything else we believe is a bit of a meme as far as implementation difficulty goes,
and for us anyway. So moving between assets doesn't just mean like a swap across chains,
it means executing swaps with specific intent or
instructions where deterministic outcomes absolutely need to be reliable ensuring data
consistency across chains where block times and finalities differ we've had challenges with this
as it leads to stale or conflicting apy or liquidity, and that can cause suboptimal allocations or slippage.
So we've looked at this and we tackle this
with the Chainlink only stack.
So we use CRE for consistent off-chain data computation,
risk checks, we use CCIP and Consero,
big shout out to Consero, for orchestration
without third-party bridges and built in
ACE, so automated compliance engine and CCID as well.
We use that for compliance.
We ensure we don't have any admin keys except an emergency pauser.
And with this, we can maintain transparency and capital efficiency.
So that's what we think some teams underestimate.
some teams underestimate.
So one of the things that LamaRisk is working on is we're developing a standardization
for POR using Chainlink CRE.
And this is a very difficult task to standardize a POR because you have token designs that have very different designs
and maybe have very different regulatory requirements
that necessitate the design decisions.
And so you may have an asset that is natively minted
on multiple chains, or it may have bridging that has burn and mint functionality
or lock and mint functionality.
So there's different types of bridging designs
that you can implement.
And what I'm getting at is this ends up
creating a nightmare with synchronizing something
that seems like it should be very simple,
like the total supply of the token across all of the environments that it exists.
Depending on how that token is designed, it could be a very straightforward process,
or it could be a nearly impossible process to synchronize that data across different on-chain environments.
different on-chain environments. And so, you know, this is also true for the more complex,
sophisticated products that we're starting to see being developed. It's becoming more common
for DeFi projects or asset issuers to create products that involve both on-chain and off-chain strategies. You may have on-chain strategies that are sourcing yield from Ethereum mainnet or from
multiple chains, and also have a centralized exchange where they're doing derivatives trading,
for example, and their asset is encompassing all of those strategies in diverse environments. That creates another level of complexity
with creating a synchronicity between all of the custodians,
the exchange sources, the on-chain ecosystems,
and compiling a robust, reliable nav
from all of those sources together.
You have problems like transfers that are in flight and settlement windows that may be
differing.
And so ensuring that synchronicity between sources is a very, very difficult task.
Wonderful. I see we've got michael up here michael do you uh want to introduce yourself i know you've had a little bit of technical difficulties
coming on up hey folks can you hear me yeah we can hear you loud and clear. Great to have you. Thank you for having me on. Yeah. Elon rugged me, apparently, I guess.
Just got a new Pixel phone and X wasn't working. So reinstalled multiple times. We're good to go.
Yeah, I've caught bits and pieces of the conversation so far.
Great to hear all these, you know, use cases from all these CRE users and adopters.
Yeah, just generally a high-level introduction and then just a few words on CRE and where we're seeing things going.
Yeah, my name is Michael Robinson.
I'm a product marketing lead here at Chainlink Labs.
And I've had the ability to work across multiple
of Chainlink standards and services,
most recently focused with CRE.
And CRE from a high level kind of vision perspective
that if you watch the Sergey and presentations at our SmartCon conference,
you would have seen similar. CRE is really how we think it is going to completely change
the landscape for how smart contracts are built, for how Web3 applications are built,
Web3 applications are built and unlock massive amounts of institutional use cases to come
live on chain into our industry. So how do we do that? Well, smart contracts started as
single chain smart contracts, right? So smart contracts on a chain that allowed you to do simple things, transfer a
token, then smart contracts advanced into what we called hybrid smart contracts, which are
contracts where you can call external data and bring data into the contract, which enabled
things like DeFi and lending.
And then smart contracts advanced further where we have cross chain and we have the ability to send tokens across chain, different apps that work across chains, etc.
And now 2025, 2026, we're entering what we have entered and we're moving forward into this space where you need even more than just data and cross chain.
You need data cross chain automation.
You need proof of reserves.
If you have a stable coin, you need that data, that asset value data.
You need nav data, that asset value data. You need all of this to happen in a secure, reliable, seamless manner. And you need a way to coordinate that in a simple and efficient way. does. So CRE is an orchestration layer that enables developers, institutions to build these
really advanced smart contracts that take things to a level that's never been possible
before. And I'd mentioned data, cross chain automation,
compliance, privacy.
These other key
capabilities are things that can be built into your
your smart contracts using CRE.
And these are things that are a requirement for institutional capital.
So it's a requirement if a major bank is going to bring capital onto blockchains,
they need these capabilities. And Chainlink is really the only platform that offers all of
these Oracle services, data oracles, cross-chain oracles, identity, etc., compliance,
in one spot that can all be seamlessly orchestrated in a secure, reliable manner with the Chainlink runtime environment.
And that's why we're so excited that it's unlocking these new use cases.
these new use cases. The early adopters like the folks on this call I think are going to be
much further advanced and have a huge competitive advantage.
We're excited to see where things go in 2026.
Awesome, thank you for that. Okay, so I know you were having a little bit of technical difficulties coming on up here. If you didn't hear Michael, the show is, as I ask these questions, it's sort of like a Roman bathhouse style. So when I throw in the question, feel free to jump in. No need to raise hands. That way we can kind of keep the conversation on high level and everybody can just keep it rolling.
So I guess I'll move on to the next one here.
For those using CRE or CCIP, how do these tools fit into your cross-system architecture? And for those not using them, what categories of tooling do you consider essential
when operating across multiple chains or environments?
Yeah, I'm happy to take this one. So within Ampli, one of the things that we picked up
was the way that agents are currently
executing transactions is a bit of a black box.
So typically, most agent systems rely on session keys. So a user would spin up a
smart account that would create a session key or a session signer key and sign that over to an agent.
That key is then stored probably on a database somewhere and the agent is then provisioned to
execute those transactions on behalf of the user, whether it's deposit funds into Morpho or
rebalance Pendle, whatever the agent is doing for the user. But that's all sort of happening centralized. And for us, that opens up
a lot of risk, especially from an institutional sort of perspective. And really what we've been
able to do with CRE is orchestrate these workflows that allow us to completely remove that requirement
to completely remove that requirement for session keys,
have agents just give intelligence
and basically propose what they want to do with the funds.
And then it runs through a validation
and a workflow on CRE that then transacts completely on chain
to then execute that on behalf of the user.
And I think one of the really cool things with CRE
is when you spin up these workflows
that your keystone forward is in the way you sort of set that up, the way you can kind
of set up your identities within those.
And the rules around that means that even though it's on-chain and people can see, okay,
wow, this is how you can actually execute something on-chain without it coming from
a valid or validated source, which would be through the don, you wouldn't actually be
able to achieve that.
And on the second half of that, we're
able to then validate that it's actually
an agent that we've given permission
to share intelligence with us for that's actually
making this request through CRE and then executes it that way.
So that's been really, really fundamental to our entire stack
and the way we built it up, which allows us to say that once you use Ampli
and you use our infra and you deploy smart accounts,
if Ampli falls away, everything will still
continue to run without the need for any sort
of centralized input.
I wanted to reflect for a second on what Michael was sharing
about the vision with CRE, and that really resonated with me.
I've thought about how with Ethereum, what really made Ethereum the golden standard for smart contract development was the EVM,
golden standard for smart contract development was the EVM,
how it opened up a development environment
for anybody around the world with an internet connection
to be able to write up a contract and deploy it
into production and make it available to the world,
and do that in a very short amount of time.
And this is really what I also see as the game changer with CRE, is it's opening up the Chainlink connectivity layer, all of the data that Chainlink already supplies, which ranges from price feeds, data feeds, HTTP calls, chain reads so there's there's all this data connectivity that you have access to within the
the cre workflow environment that you can use to create new new categories of applications
that are externally aware to to an extent and level of efficiency that that hasn't been possible
before um so what we're uh what we're building is um right now is islam. So what we're building right now is LamaGuard NAV.
We're building it for Aave Verizon,
and it's risk-managed price feeds for regulated RWA assets
that have a single source of truth,
which is the asset issuer delivering the price value. And we have
to have risk management around safely integrating that price data into DeFi applications like Aave
Horizon. So what we're able to do with CRE is create a risk managed price feed, deliver that
data on chain, which imposes our own bounding logic
that we run within CRE. And we also deliver on chain secondary data points, which is data about
if the bounds have been breached by the asset issuer's nav report. And if so, we identify that
as an anomaly that needs to have some kind of an emergency action.
So we give the DeFi application external awareness about our opinion of if there is an anomalous event that's warranting some kind of emergency response.
And the great thing about what we're able to do with this is that this solution, although we're developing it for a specific use case with Aave Horizon, it's actually extensible to scale the types of custom solutions that we offer so that
they can be generalized and integrated by any chain, any DeFi protocol.
Wonderful.
Michael, I've got a question for you.
So what's the easiest way
for a builder to start utilizing
these tools or to start building
with these tools?
Easiest way?
You know what? We just did a CRE
masterclass with Coinbase and their X402
agent. I would check that out. It's on YouTube. If you just go to Chainlink's channel, go
to our live section, it should be one of the most recent videos. And that's a great video
to hands on keyboard, walk people through how to get started with CRE and, you know, build a cool use case.
We have a bunch of different templates and you can follow in our docs to get started and check
it out. And yeah, see what, see what you can build. I think a lot of the feedback we've been hearing is
that the use cases being built with CRE,, they wouldn't be possible without CRE.
So it's a,
it's a pretty cool position to be in and that there's really not even a
competitor that can offer the same thing that, so this is, this is your,
your environment to build something that's that hasn't been possible ever
before, you know, be creative. We have all sorts of use cases.
If you check out the CRE launch blog or check out our website,
chain.link slash CRE, you can see some various different use cases.
And yeah, I highly recommend everyone.
If you're listening in CRE.chain.link, sign up,
play around with the, the runtime and yeah just get your
hands in there and and uh see what you can what you can build awesome yeah thank you for sharing
for that and i want to welcome up rock rocks come up here um did you want to introduce yourself rocker sure sure hey guys good to see you all
I'm out here in Las Vegas for CES at the moment over actually at one of Michael Turpin's house
houses from bid angels and yesterday we had a couple investor events here,
a bit of an investor summit.
So that was nice to see different kind of family offices
and others in the traditional markets
who are actually all quite interested in crypto.
I think that was pretty interesting to see.
You know, a lot of these people
have very little experience in crypto,
but they're very eager to learn what's happening in our industry,
how do they get exposure.
And actually something a bit off topic here,
but I think might be interesting for many of us in this industry
that are maybe looking for some hopium.
But the kind of narrative between all these these traditional
investors family offices there's a lot of silicon valley folks uh but the the kind of theme amongst
a lot of them that was that they saw and and in their circles they're all talking about that this
is kind of in in different industries you have these adoption curves and you have kind of this overshooting hump where there's a lot of investment in a certain technology, maybe overinvestment and excitement.
And then it kind of pulls back and then you have a lot of building and then you have a massive growth phase after that, like kind of an S curve.
massive growth phase after that like kind of an s curve um and they were all kind of saying similar
things which is that they think we're in the a bottom part of a hump getting prepared to go into
the s curve and so they all were trying to figure out how do they get exposure to our industry
and it was this really common theme amongst them so i thought that was kind of interesting um yeah
so yeah good to be here.
Really cool to see what you guys are building at Chainlink
and all the teams that are building on top of it.
Yeah, I've been digging into CRE quite a bit,
looking into this, especially now leading up to this show.
And yeah, it seems really interesting.
It kind of feels like the more I look into it,
the more I realize, wow, our industry is still pretty immature in terms of tooling.
And it's kind of surprising that we don't already have things like this.
I know that this has been around for, I don't know, a handful of months now.
When did CRE actually go live?
Early November.
Yeah, so SmartCon, November 5th is when we officially went live.
There were some early access users prior.
But yeah, it's really only been a couple months here.
Wow.
Yeah, super cool.
Yeah, seeing what it does and the capabilities and then hearing what you guys are talking about here. I've been listening for the last like 45 minutes, just trying to wake up. I was up late at this investor event. But yeah, seeing what this opens up, I'm just kind of surprised thinking to myself, because I was there for the early internet. And I'm realizing, you know, hearing people talk about this. Did you just say you were there for the early internet and I'm realizing um you know hearing people talk
about you said you were there for the early internet yeah okay yeah I was uh I was selling
actually digital currencies in the late 90s before Bitcoin in video games to Brock Pierce and Jonathan Yantis, who later made, you know, Tether and EOS
and DNA Fund and Wax.
But yeah, you know, seeing the early internet be built
and the primitive tooling we had,
it kind of feels like crypto, seeing this come out
and seeing what this opens up and then realizing wait a second we
didn't already have stuff like that uh you know with all the infra there's like a ton of infra
providers out there but we didn't already have something like this is kind of surprising
um and it kind of made me think you know maybe this is why we don't have you know massive
mainstream adoption already is because we just didn't
have simple tools like this to make orchestration and coordination across chains and protocols
and oracles and feeds and all these things.
We didn't have a tool to make all this just work.
And now we do, thanks to Chainlink.
So kudos to you guys.
Robert, I might add some color to what you said from our corner of the industry here
at the intersection of agents and blockchain.
I find it fascinating how it almost seems like
we built blockchain infrastructure
before its actual ideal user was on the planet,
which is agents, in my opinion.
And there's this interesting synergy
where agents require blockchains
in order to, let's say, transact value,
perceive reality, have a temper-proof
and human middleman-free way of interacting with the world.
And at the same time, agents make blockchains usable
and solve the terrible UX.
You could argue that blockchains were never
great for the regular human user, right?
It takes a lot of work.
And now we can use agents to abstract a lot of that away and you know using all that infrastructure and making that
accessible that you just described and I think that's a super exciting time
right now
yeah for me CRE it actually kind of feels like the whole story of Docker and Kubernetes, how you had Docker and containerized applications and services around 2012, 2013.
And it wasn't until 2016, maybe 2018, where Kubernetes really kind of liberated the idea of using containers
in mass to do massive orchestration of services.
It feels like CRE is almost at that crux point of,
or the pinnacle of how Kubernetes was able
to change container orchestration.
I think CRE is to smart contracts,
what Kubernetes might be to Docker containers and the like.
So when it comes to like the speed of progress, I think it's good to keep things in perspective
that we're an industry that moves at 100 miles an hour. We're constantly innovating,
and that's thanks to being an open source ecosystem
that is just extremely inviting for builders
to come in and experiment with things.
But that has historically also resulted
in huge amounts of value being lost due to hacks or other kinds of exploits.
And so, in a way, the years leading up to this moment right now,
and the past, let's say, seven years or so since Chainlink's been on mainnet,
Chainlink's been on mainnet, it's been really a practice of slowly grinding our proof to the
world, to the financial system generally, that what we have to offer with this blockchain ecosystem
is indeed the future. And so I feel that we've been hoping for many years about this concept of the convergence
of TradFi and DeFi. And it's always been sort of like a distant goal. But I feel that in 2025,
this was the first time where we've really seen a true pathway where for me, I can see the vision
of how this convergence happens is within sight.
And I think it's very timely for CRE to be being deployed onto mainnet at this moment
where Chainlink has been establishing its gold standard reputation for many years now.
And it's now opening up this new platform, CRE, that allows any developer to come create custom solutions using its proven data infrastructure.
data infrastructure.
I'm curious, as it's only been out for a few months now, are we seeing any
institutions adopting this early?
Or is it still just kind of startup mode for this CRE stack?
I can speak from our end.
We did the, Struxie did the gold-backed Toga security for PMUSD. The company that owns that underlying gold security is a publicly traded company.
The company that owns that underlying gold security is a publicly traded company.
And we use CRE to tie the off-chain artifacts and off-chain attestations into an on-chain attestation for that gold.
So from our perspective, or from my perspective, that's an institutional adoption data point.
what is the uh how does that work so the price is still um maintained just like uh any stable
How does that work?
coin or anyone else that would want to pay something to some asset value but this is more
of like um just showing on-chain audit trail or yeah on-chain audit trail as well as well as
driving the math by behind how many tokens should be issued at any given time.
So one of the tricks with gold is that I didn't realize until going down this exercise is you
kind of think of gold as like this thing that sits in a vault somewhere and you're going to just have
access to some reporting number every month and it's as easy as flipping that onto the blockchain.
But what you really end up finding is that there's gold and gold equivalents
and gold can exist under the ground,
gold can exist in ore state,
gold can exist in refined in a vault.
So there's all sorts of math and all sorts of discounting
that has to be applied to the underlying gold
on the balance sheet.
And what you end up finding
is you're actually representing that balance sheet
or the profit and loss statement on chain
rather than the gold bars.
So what that comes like, if the gold is under the ground,
you can imagine that there's PDF artifacts that come in
from some company that's done geological scans.
And we have to have professionals audit those documents,
make sure that we all to have professionals audit those documents make sure they're real
pull data out of those documents so that the data ties to what's on their what's on their financials
you know when they've been audited so you end up finding that cre is really useful for collating
all of this information and piping that onto the blockchain uh if and when they're ready to do it token issuance.
Wow, I didn't know that when, you know, for, I guess, PAX Gold or USD Gold
or Tether Gold or whatever it is,
or any of these that, what was the one you mentioned?
Which company?
Yeah, USD, it's actually, it's a new stable coin
that's being, getting adopted more recently in the Curve ecosystem.
It's deployed by Rack.io, but we're the underlying technology arm that's interfacing with the publicly traded company that owns the gold asset.
And the gold, is it pegged to the price of gold or you said stable coin,
is it pegged to dollars or something?
Yeah. The, the, the stable coin is derived from the, the,
the underlying token, which is pegged to gold.
And we use CRE and price speeds to do some,
some map underline to make sure that,
that those tokens are atomically represented and
staked and locked in the right positions at all times when you say stable coin do you mean
pegged to the price of the dollar or pig sorry pm usd is pegged to the price of the dollar
uh but we also maintain a certain number of ounces uh that can be that can be minted into stable coins that's pretty badass that's pretty cool
um i'm just trying to think of what the how that gosh a lot of questions there on how the math
works out and what you know if you have a billion dollars of gold uh how much stable coin do you
have and does it account for volatility is it over collateralized and things like that yeah it can get it can get complicated really uh really quickly happy to
do a follow-up case study or something especially when you're counting not just gold in vault but
also like and refined gold but also underground gold um futures contract is also their goal
like it's really gold and gold equivalents and of course
you can say the same for silver or palladium or any precious metal here very interesting
so that's uh that's pretty cool though that i mean you know this industry never ceases to amaze me
the innovations we come up with that weren't really possible in TradFi.
And we are just constantly inventing things like perpetuals, flash loans,
and all kinds of weird, awesome new tools that the world will want to adopt.
But yeah, that's a cool example of what CRE is capable of doing.
Yeah, just to follow up your question earlier, who have raised that institutional adoption.
So if you just check out Chainlink runtime environment launch blog from November 5th,
you'll see a ton of these. But Connexus by JP Morgan and Ondo did a DBP transaction
between public and private chains powered by CRE.
Swift, Euroclear, and a bunch of institutions use CRE
to streamline this new process for corporate actions.
Swift and UBS leveraged CRE to connect existing systems
with blockchains and then complete transfer
of a tokenized fund.
CRE with Swapper and MasterCard obviously,
and yeah, a whole number of other major use cases.
So CRE has seen adoption across TradFi institutions, DeFi, even Web2.
There's a demo like I mentioned with Coinbase, but also with Google Cloud and AWS using Google
AI to create a prediction market using CRE. So the use cases are honestly endless
and the adoption's been pretty amazing across the board from different types of
businesses, protocols, etc. I have a hard stop,
but I want to thank you for having me on here.
And thanks to our users as well
for all the valuable feedback
and for building with CRE and Chainlink.
And thank you to the community for listening.
And I'm going to step out,
but let you guys keep keeping the conversation going.
Thank you very much.
Cheers, Michael.
And good to see you.
And everybody give Michael a follow.
He's definitely shaking and moving things in the industry uh yeah good to see a bunch of you here uh austin
good to see you also ash zen etc um and patrick i like your cowboy bebop uh pfp is that what that is thank you yes great taste nice yeah actually uh cindy uh my girlfriend
did the casting for the live action on netflix of that wow okay amazing yeah that's that's awesome
yeah it's a it's a great cult classic um cool, guys. So, yeah, I wonder how does Chainlink monetize this?
So you guys built this amazing orchestration layer,
and people are using it,
and there's all kinds of new things that it opens up
for more institutional-grade products.
And did you guys just build this altruistically to make the industry move forward?
And if the industry moves forward, Chainlink gains more valuation or adoption
or Oracle feeds or that kind of stuff?
Or is there a way you guys directly monetize this product?
Well, I can talk a little bit more about the specific work
that we're doing with Aave Horizon.
And this goes back to what we've been talking about,
this convergence of TradFi and DeFi.
So anybody who's not familiar with Aave Horizon,
it is the largest RWA lending platform,
and it uses regulated RWAs as collateral to borrow stablecoins. And these are right now,
everything that's listed on Aave Horizon, these are regulated to the extent that they require
KYC, AML, and so they have whitelisting requirements.
This makes it particularly difficult
to onboard these types of assets as collateral.
So as we're seeing more demand for RWA assets coming on chain,
we're trying to find ways to safely integrate them
into DeFi so they have more utility
so that we can drive more adoption.
And specifically, what we're doing for Aave Horizon integrate them into DeFi so they have more utility so that we can drive more adoption.
And specifically what we're doing for Aave Horizon is building risk managed data feeds
for the collaterals that are listed onto the platform.
So these range from like T-bill funds, corporate debt, CLOs, soon we're going to have public
equities that are going to be listed on the platform.
And the kind of the business model that we have is that we are a risk manager that have a partnership with the DeFi protocols.
We work closely with the development team on Aave Horizon, which is Aave Labs.
on Aave Horizon, which is Aave Labs.
And we present ourselves as a independent voice
to help them make good sustainable solutions
and to implement solutions like LamaGuard NAV,
which is built on Chainlink CRE.
So this is currently in production
in its first version right now.
And it's currently built in a way
that we collaborate with Chainlink Labs to deploy
an external adapter that implements our risk management methodologies for safely delivering
that NAV pricing on chain.
But we're working right now actively to implement Chainlink CRE as an upgraded version with
additional risk management capabilities built into it.
That is to say, an emergency response to freeze markets when some sort of anomaly has been detected.
And for chain link side, the business model is simply you need to pay to run the workflows. So it requires LinkToken to pay for the operation of this.
And since we have a partnership with the protocol team,
we have a mandate to deliver the value to them in any way that we can.
And we pass through the costs to operate the
infrastructure that they need to run their system.
I think just to add something there.
So on the backend, everything's powered by Lync.
But for users that interact with CRE, so CRE is a UI, it's an SDK, it's a CLI. And so the cost comes from compute and the compute is
basically paid for in link, but users have the ability to basically use USDC and soon even just
like plug in their credit cards. And so like this will like lower the barrier of entry for folks
that, you know, aren't used of purchasing gas but
want to build an application and deploy an application that is connected to like any
off-chain cloud system and execute on like the decentralized blockchains that we are
all working with so what we see are a lot of new use cases coming at this intersection of instead of deploying code on an off-chain system or off-chain database and then on the blockchain, being able to deploy institutional grade hybrid pieces of code that natively connect these two worlds.
Naturally, we see a lot of new use cases for computes and this will all be value accrued back to the link token.
Nice.
So what has been the biggest breakthrough for CRE
that allowed you to scale without worrying
about cross-chain interoperability?
So if you look at Chainlink, you have data across chain compliance,
identity, you have automations functions, you have a lot of products that weren't composed as
composable with each other as before. And so initially, we actually built CRE as a solution
for us. And this solution that we started to use became very appealing to institutions,
especially because it's programmable
in like TypeScript or Go.
You don't have to learn Rust or Solidity, et cetera, right?
And so the real breakthrough came
from like a cross-chain perspective
where any chain that CCIP is on
is natively a destination of any endpoint where aRE, a user building on CRE wants to connect
to. So we've actually built this on top of everything on the interoperability side. So
today CRE is deployed on a few chains, but soon it'll be basically everywhere where CCIP is.
And the lift for us to do that because a lot of the groundwork that went into building CCIP is and the lift for us to do that because a lot of the groundwork that went
into building CCIP and the interoperability stack is now being able to deploy CRE on top
of that basically. Awesome.
Okay.
Does anybody else have anything to build on top of that
or that they want to add to it?
Yeah, a couple of very brief points.
So we used CRE to apply our parent child architecture, uh, for managing
cross chain states, um, where a single parent, um, uh, chain track system-wide
states for all other child.
So that was incredibly useful.
And something else, which hasn't been mentioned is CCT.
So the cross chain token standard.
So we baked a CCT standard into our yield coin share token, basically because it was
a very easy way of extending its reach to multiple blockchains with minimal effort.
And we've avoided loads of complexities associated with traditional bridging
solutions.
And in the end, this is going this is what is going to fundamentally allow users
to redeem Yieldcoin from any chain, any time.
So yeah, I just thought I'd add that.
So what are the core cross-chain reliability or timing problems
you've had to insure in year-round to make this work safely.
Is that for me, yeah?
Yeah, that's for you.
Yeah, you're a judge, right?
Yeah, that's right.
I thought that was you, yeah.
So we've had a couple of issues with timing.
So we had some reliability issues for inconsistent data feeds
across chains, so that led to varying block times, um, and stale APYs.
That was a bit of a challenge.
So timing, um, again, latency in rebalancing.
So if we use manual or event-based systems that could take minutes or hours.
Um, like if you're trying to do cross asset swaps, that causes insane amounts of slippage,
especially during volatile shifts.
So yeah, that's basically what we've had to deal with recently.
Nice.
And I know that you've said that,
I think if I can refer back,
I think in the past you've said that CREs helped you reduce your smart contract complexity by over like 50%. Is that right?
Oh, yeah, yeah. Crikey, yeah. So basically CRE is an amazing Swiss...
Crikey.
Yeah.
It reduced our smart contract complexity by about 53%.
It allowed us to offload heavy computations, decentralized workflows
that were previously incredibly bloated on chain.
The biggest impacts came from architectural changes.
So replacing AWS intermediaries and DeFi Lama API calls.
We replaced this with CRE because it's tamper-resistant, does data fetching and filtering.
It eliminated quite a lot of unnecessary on-chain calls and reduced the APY impact calculator
that we've got by about 100 lines of code.
On top of that, shifting rebalances triggers from event listening via...
So we shifted it from Chainlink automation to the API. On top of that, shifting rebalances triggers from event listening via...
So we shifted it from Chainlink Automation to CRE, and this simplified the rebalancing
contract by removing multi-step orchestrations and enabling second fast.
So like triggers are triggered in seconds by a Cancero.
So we can do things at retail speeds.
Third on top of that, abstractions like CRE's off-chain environment. We use that for complex APY risk calculations because it's incredibly infeasible on-chain. It's too expensive,
sometimes impossible to do. It's basically costly.
CRE is amazing for compliance hooks, so we can use ACE for KYC, AML, CCID,
and that helped modularize our vault design.
So that means we can accept any protocol, any chain, and any token into our system. And this is all done via minimal audits and redeploying.
So overall, this created an immutable open source system with no admin keys,
focused on on-chain logic only.
And CRE handles the computational heavy lifting for scalable, secure
cross-chain operations.
It's an absolute beast, CRE.
We absolutely love it.
So yeah, those three key architectural changes were the things that, that really
helped us the data fetching and filtering the off chain, uh, rebalance triggers and
orchestration and abstractions for complex calculations.
And so these chains made us real resulted in a leaner, more scalable architecture.
Um, our on-chain logic now emphasizes verification and execution.
Well, CRE basically provides the intelligence layer.
So our Chainlink native stack combines CRE, CCIP, and Cancero to deliver
institutional grade safety.
We are going to be displaying on our website, the APY's net of fees to ensure
transparency, and we can essentially expand to thousands of chains. website, the APY's net of fees to ensure transparency.
And we can essentially expand to thousands of chains, any chain we can get into.
So we're trying to position yield coin as a default yield optimizer for initially
stable coins, for looking at low risk DeFi by placing the stable coins in lending
markets like Arve and Compound, things like that.
And eventually we'll move this to any token, any protocol, and any chain.
So yeah, Chainlink has really helped us moving forward.
It's reduced quite a lot of stuff and made things really easy,
reduced costs, reduced time, reduced complexity.
It's absolutely brilliant.
So yeah, I love CRE.
So yeah, hope I got that across.
Yeah, you did.
So when you mentioned earlier DeFi Llama,
like API or feed or something,
is this just how DeFi Llama gathers info
from the blockchain to present on their site?
The DeFi Llama API is the most gargantuan API we've ever come across.
It gives you data on all sorts of shenanigans going on in the blockchain world.
So what we're able to do is look at a CRE-based system that can help with our APY calculations.
I think we're going to be using DeFi Llama as a backup and go to the chains,
the protocols individually for their APIs.
But again, this depends on latency.
This depends on finality.
So yeah, CRE has helped us a significant amount getting us a bit further,
but we're going to use multiple places to get the APIy data essentially you're saying when you guys are looking um for data you used to go to
defy llama and but defy llama is as is uh not as efficient or easy to access or too complicated or
too much data or something so now instead you using CRE to get some of that data?
Is that what you're saying?
Yeah, that's it.
I think we're comparing it at the moment.
So we can go to Aave, we can go to Compound,
we can get their APY data,
and we can also go to DeFi Lama as well as a centralized point.
Oh, interesting.
I wonder, guys, are there any
for you as a builder or for Chainlink
as the provider of this
it sounds great
sounds like the best thing since sliced bread
but is there any drawbacks
to this? Is there any
centralization risk
or is there
what are the either risks or
negatives or trade-offs of using this versus
what we had before? So we've tried right so I don't really care about chains or centralization
or things like that I care a lot about decentralization so we're trying to remove any
centralized points of failure throughout this so initially we had a centralized point of failure, which is an AWS server.
We took the APY data and we scrapped all that.
Um, and now we just use chain link for CRE when it comes to things,
holding us, uh, to, to things.
No, I I'm not sure that, uh, I think we're, we're creating a system, which
is far better than anything we've had before.
And CRE enables us to do that so that we can come out and say, look, we've built
this thing, here's all of our data.
We can prove everything's appropriate.
We can rebalance appropriately.
Um, we can make sure that slippage is taken into consideration.
Liquidity is taken into consideration.
Gas costs are taken into consideration and we can do this all through CRE.
So if we were to do it before using automate automations and functions
again from chain link, it would, it, it would have cost a little bit more.
It would have been a bit slow, and we would
have to rely on some on-train metrics.
Yeah, I think we're building a better system.
So in your opinion, this further decentralizes things.
But given that it is one partner, could this be a central, you know,
point of failure where like, let's say it, you know, the risk of people that people say about
like, in FURA or things, if CRE, for whatever reason, stopped working, I don't know,
wherever their chain links incorporated, if that government like told them one day,
where their chain links incorporated
if that government like told them one day,
hey, freeze all operations, for example,
or anything like that, what would happen?
Okay, so we're using CRE to query on-chain data
for each protocol and also API fetching
from DeFi Llama, API integrations of APIs, essentially.
And we're computing it ourselves.
So just to make sure that we're getting the right data.
I think it's worth calling out that CRE
is running on top of the decentralized Oracle network.
So there could be hundreds of nodes
sitting under this helping fulfill work.
So the idea that that part is actually not centralized,
so there is no going down from my perspective or from my point of view.
Oh, very cool. Very cool.
Yeah. So from our perspective,
so it's not it's not Chainlink just running, you know,
this or providing this feed from their website.
And then if it goes down, all these other little pieces go down.
And this is part of the the chain link uh validator network we're we're inheriting yeah the the robust uh battle
tested chain link infrastructure and this was this was very important for us when when we were
developing a solution we wanted to make sure that um llama risk would not be a central point of failure at any point in the process, because
what we're trying to do is deliver safe pricing data on-chain.
And this is a critical dependency.
If the pricing data fails for some reason, that means that DeFi protocols that use that
data, there could be bad debt that's assumed to the lending protocol.
It could wrongfully liquidate users.
So this isn't something that we want to have direct controls over.
We want to establish a system that can run autonomously.
And so therefore, when we develop a workflow on Chainlink CRE, we implement the code similar to how we might implement smart contract code.
And that has a unique identifier attached to the workflow code.
So when that data is pushed on chain, you can verify that the workflow we say is running
is the one that's running.
And any update that is made to what workflow is being run, that is, if you wanted to upgrade the code
of the workflow that CRE is running, then we delegate that management to the protocol team
that we're working with. So it's their responsibility to actually execute any upgrade
that happens. So nowhere in the system for our risk-managed price feeds can LamaRisk arbitrarily go in and say, well, I'm going to change the workflow and introduce some different logic, or I'm going to stop the workflow from running, for example.
Because, like I mentioned, these are critical operations that are directly responsible for securing large amounts of value.
On Aave Horizon, there's over $ large amounts of value on Aave Horizon.
There's over $600 million of value right now, and it would be completely unacceptable for us
and not what we would expect from users that they would be willing to just assume that
the Lomeris backend running on AWS is going to continue running reliably and honestly.
So that's why we delegate the data infrastructure part of our system to Chainlink infrastructure,
which is already battle-tested over many years, and we implement our own custom logic
that we work very closely with the protocol teams who are ingesting it to make sure that it's meeting their standards.
Makes sense. Cool. We're getting pretty deep in, I think the the tech weeds and the security weeds on all this stuff
which I think maybe average people might struggle with some of this but it is it is pretty cool to
see all the stuff being built in that our industry is maturing and we're having a lot more kind of Lindy effect.
And thanks to tooling like this,
LamaRisk and all the tooling, Chainlink's building and others here on stage and in our industry in general,
that we're seeing a lot less hacks, which is nice.
In the earlier days,
yeah, it was like we saw new hacks every other day, but it's definitely, I think the industry is maturing.
And I think people are feeling a lot more safe using a lot of these protocols because they're battle tested, because of Lindy effect, because there's different redundancies and security measures.
Yeah, there's actually been a trend over the past two years or so since 2024, where there's been quite a widening divergence between increasing TVLs in DeFi and reducing losses due to some kind of hack exploit.
So over the past two years, there's definitely been a maturation of the industry that I think we've learned from past mistakes, many of the past mistakes, and we're surely going to be faced with new ones as we continue becoming more sophisticated with the types of financial products that are being created.
But there's definitely been a positive trend over the past
couple of years. But maybe another thing that I wanted to talk about in terms of just general
trends that we're observing in DeFi is that over this past year, we've seen the types of DeFi
products that have gained most traction. Well, one of the big narratives this
past year has been leverage looped lending. So these are lend markets where you have like-kind
assets. And what has been particularly popular are yield-bearing stablecoins and, by extension, Pendle token stablecoin markets. So that's fixed
rate yield that borrowers can earn. And through this leverage strategy, they gain maximal yield
exposure to their collateral and earning the spread between the rate that they're earning
on the underlying versus the borrowing costs that they have.
This trend, which by the way, like Aave now is backed by a significant, something like
a third of its TVL is backed by these types of strategies.
Many of these strategies from a product perspective, there's been a trend towards pegged pricing.
So traditionally, we've relied almost exclusively on using secondary market pricing to determine
when liquidations should happen. But more recently, over the past year, there's been
a massive shift towards pegged pricing because from a product perspective, these leverage loop
markets are much safer for borrowers if you're not
liquidating them on assets that may be susceptible to manipulation or transient
dislocations in their pricing. But we consider this to be a very intermediary solution because
you're moving the credit risk, the solvency risk onto the lenders and onto the DeFi protocol that is serving these markets.
And that's one of the main reasons we're putting a lot of attention right now on risk-managed
pricing because we think that the next phase of sophistication is going to require us to use
solvency stress signals to be able to determine when some sort of risk management action is required.
If we are choosing not to use secondary markets as our main signal for fair market value, then we need to incorporate other signals,
like if there are underlying asset dependencies that the collateral has that we can get signals from,
if there's an observable redemption on-chain, for instance, that we can view redemption buffers,
if there is a robust POR that we can use to determine collateralization of the asset,
those kind of things can be incorporated into risk signals that we can use to create next generation pricing for assets. And we can do that all building on
top of Chainlink CRE.
Very cool. Yeah, that's interesting that you say that because, you know, earlier when
I said we're
seeing a lot less hacks we've seen a lot less hacks but we have seen still quite a bit of
manipulation where people are messing with you know uh trying to either do things that
traditional markets would do like a short squeeze or uh long squeeze or uh trying to
uh get people liquidated and and these kind of things.
So, yeah, all the stuff you're talking about is different ways of thinking about
how do we stop these bad actors from trying to play these games
where they manipulate assets to harm the users
who are just trying to either borrow against their assets in a safe way or, or lending get a yield.
Yeah, and maybe a good exampleance really, experienced extreme price dislocation
in the wake of market-wide volatility.
And it was because Binance had used its own internal pricing mechanism for USDE, its own
internal oracle, that it was mispricing that asset, which was used in leveraged positions.
And there was a deleveraging cascade that took place that temporarily moved the USDE,
which is intended to be a stablecoin, down to something like $0.65 on the dollar temporarily.
something like 65 cents on the dollar temporarily.
And on DeFi platforms like Aave,
we've previously implemented safeguards
to prevent that kind of manipulation
from wrongfully liquidating users.
Although it's an important point to make
that the Chainlink feed that was serving this data
was remarkably resilient during that event.
And the price that was delivered by Chainlink deviated only slightly.
But it's a good example of just what can go wrong and how quickly we can be taken by surprise.
When volatility strikes, it be uh swift and extreme
it's kind of rough though because are there pros and cons there i mean in a way if an asset
okay if something's being manipulated and there's nefarious actors doing it
for just like quick gains sure we all would probably agree that's that's not
ideal um there may be some people who would even make an argument that that is fair and like it's
like kind of like code is law kind of stuff right but um i think most of us would probably say hey
we don't want people manipulating and trying to liquidate people just because they see an opportunity. But at the same time, if something,
if a price is moving quickly,
because I'm thinking while you're saying that,
well, could you have something like,
instead of liquidating based off low latency,
you purposely increase the latency,
you have like a rolling, you know, a 24-hour average.
Yeah, you definitely...
That's definitely something, if I understand your point correctly,
that you can create custom logic on top of how you're deriving a pricing.
So like...
Well, but my... Sorry my someone's calling me, but my final point is,
but isn't there a world where you kind of need that really low latency and you need,
if you were a, if you're a, a, a lender and, or if you're anyone in this, in this chain of events
that you don't want to be lagged behind because if there's
something happening, you need to unwind your position quickly to protect yourself.
So are there negatives for if, or are there other kind of events now where if we make a system that
doesn't trigger super fast on these manipulation events, that it'll also not trigger in an important actual downturn
where people need to unwind and close out positions to de-risk.
I think that is less about speed and more so about, you know,
the job of an oracle is to get as close to the quote-unquote truth as possible.
So I think you want a robust signal
that is made up of the most significant markets
that give the signal.
So in case of that Binance example
that Wim Hof mentioned,
that was, I think it was also the case
on several assets and WBTC as well.
They were all used as collateral
on that platform for margin positions.
So that was where that cascade started.
And the issue wasn't necessarily the speed.
It was more so that they weren't factoring in any other places where these assets traded.
I believe that the main curve pool for USD only barely moved during all of these events.
So having that in the mix would have already helped liquidation cascade happening.
So having more...
That was going on there, by the way.
Wow.
So having more coordination
between all of these different sources of data
and price,
maybe something, who knows, like a coordination layer like CRE would help in these situations.
So they're not pricing off of just one price area that is subject to manipulation.
So, kind of what you're talking about with the trade-offs that one might make and how
they're determining the pricing, depending on factors like what sort of diversity of
venues are you considering in your pricing?
Are you applying some sort of custom logic that is arbitrarily slowing the amount that
that can change?
Are you using other risk signals aside from secondary market that is assessing the fair value and using that to determine when liquidation is needed?
So in general, these decision makings could be unique to each individual protocol.
There could be just custom decisions based on the type of asset that is the underlying and the, I guess,
the priorities, you could say, of different DeFi protocols. And overall, this whole category of
decision making has been described as risk oracles. And risk oracles is sort of a very
general term of doing operations that really Chainlink has already been doing as long
as it's been on mainnet, which is making certain decisions about how do we determine what is fair
market value, for instance, if you're using cryptocurrency prices or asset prices and using,
you know, what kind of methodology do you use for that, like a volume
weighted asset pricing, a median pricing between venues, something like this. And with the
introduction of CRE, this is really opening up this concept of what a risk oracle is to the world,
to risk managers or to DeFi protocols that want
to implement their own custom answer to this question of what trade-offs am I acceptable
with?
And so what has been a previously unified product of opinionated and unopinionated data
to deliver pricing, now it's something that Chainlink with CRE
is managing the data infrastructure.
And it's saying, hey, DeFi protocols,
risk managers in the space,
build your own opinionated decision
on how do we derive fair pricing
for this particular use case,
which I think is a really great way for Chainlink to scale
as we move into a world where there can be thousands of assets
across hundreds of chains and allowing a space
where that customization can take place
for all of the different types of categories of assets
that are already online and many more that will be coming in the near future.
Amazing.
So Ryder or Darren, do we have any more questions
or does anyone in the panel want to jump in there?
We do have some questions,
but unless somebody wants to jump in there,
I do have a question for Swapper Finance.
Is somebody from Swapper still here?
Chris looks like you dropped down.
Okay, I can take this question
on over to
oh wow, I'm having a hard time saying this name.
So Instruxy?
Or just Austin? Are you here, Austin?
Yes, I'm here.
Okay. Awesome. Awesome. All right. So Instruxy works at the intersection of op-chain systems and on-chain enforcement. What are the biggest challenges you see in reliability translating roles,
permissions, and data between these environments?
Yeah, it's really kind of architecting an environment where non-crypto-native individuals can participate safely.
Because at the end of the day, most of the people that are holding these off-chain assets
or building these off-chain systems, like, I don't know, a Bricks Vault API, for instance,
these are all non-Web3 native individuals.
So kind of architecting a system where a multi-sig exists,
training these individuals on how to use a wallet,
building systems that use things like Privy with policy,
you know, native policy management on the wall itself. Um,
I find that to be a kind of a bigger challenge than the actual middleware
development. Now that we have CRE, a lot of the heavy lifting, um,
kind of takes care of itself. It feels almost like you're interacting with,
uh, you know,
a serverless cloud flare component as a developer or AWS Lambda function nowadays.
So from a developer's perspective,
I've found the middleware development exercise
to be the easiest and so actually projecting that
into a world where these non-native users
can take land securely, give us the data
that they need to give us
to be successful but also architecting a way for them to do the signing procedures
token issuance things like that um so yeah
all right does anybody have yeah just anything to say on that?
Yeah, just to follow up on that. So, you mentioned proof of reserves. Like, how much easier does CRE make that? Like, what that would look like. Let's suppose you had a database and it has the invoices
or it has the data that you need
and you're able to sum down the attributes of that database
and you arrive at $120 million should be put on chain.
Well, you'd have to go to Chainlink
and you'd have to build an external adapter,
is what they call it.
And you take that external adapter,
you bring it to Chainlink and you put it into a system or a backlog.
And that piece of, that microservice that you've developed
has to be installed, so to speak, at different node providers.
So this was a process that maybe took anywhere between three to five weeks, depending on
the backlog or even the complexity of the microservice that you would have to develop
to get to that data.
Well, nowadays, everything's packaged up in a nice SDK and you have an abstraction
that feels more like you're,
it's an abstraction layer written in Go or TypeScript.
And it actually feels like you're writing a CRE job
or you're actually building a proof of reserve
as opposed to adapting like TypeScript
and building a microservice in a Docker container.
So it felt like before this,
you were building on old infrastructure
or kind of the way you would do this,
the old coding way,
on top of the lead time required
for you to get that microservice
like deployed into the node environments,
made it much more complex
than just building directly on the CRES today.
So now with a little help from AI and some of the templates that the guys have put out,
I mean, you can put a proof of reserve on it over your lunch break. So it's gone from hours and day,
it's gone from many weeks to just hours or days to get something safe
and secure out the door.
Wow, that's pretty cool.
Which is kind of nice to hear,
because I know in the past with other parts of Chainlink
in the earlier days, I know it's been getting a lot faster now, but I remember, you know, talking to Chainlink for getting oracles on a newer chain.
And even with paying the, you know, at the time pretty high cost, it was still like, yeah, we're at least six months out, you know.
So getting, being able to deploy these things a lot faster is definitely great for the industry.
I think serverless is a great metaphor because if you're, if you've developed in the web
two space before this, you can imagine that let's say you build a Python server or a TypeScript
server or a Golang server.
You might have to go to a data center or go to AWS and you take out an EC2 instance,
which is just a VM, then the VM has
to have a network wired up to it. Then you have to go actually install your, your compile your
program, install that program and run it in a, using a CICD pipeline. And there's all of these
like incremental steps to actually getting your application hosted in a data center or a VM environment. Whereas with AWS Lambda
or Google Functions or Cloudflare workers, these are serverless components and you simply
write code using their guardrails and their deployment package and you just push code
into their infrastructure and they handle the hosting
and the networking and all of the all of the complicated things that you you would have had
to do in the past those abstraction layers now handle that infrastructure for you so for me
the the closest uh metaphor for like for a developer is we're relating something like AWS Lambda and other serverless technology to the old way of deploying server components.
Whereas now CRE is very, it feels like a serverless experience.
Okay, awesome. Yeah, thank you for that.
So the next question, I'm'm gonna shift it over to amplify
i see we still got patrick here uh with this i'm still here yeah cool so agent validation
without centralization is a difficult problem uh what makes enforcing rules and policies across
agents wallets and multiple chains uniquely complex?
Could you repeat the question?
I didn't hear the first part, Richard.
Sorry, I'm curious. So agent validation without centralization is a difficult problem.
What makes enforcing rules and policies across agents, wallets,
and multiple chains uniquely complex?
Yeah, so what makes this complex today
and before we build what we built
is that your intelligence from the agents
and your execution on chain and the whole permission
management is very, very entangled with each other
and intertwined.
So the main complexity we had to solve was figuring out a clean separation
there between where the agent lives, how that works, and the user that still
needs to own the full transaction building and never break that self-custody.
So that was like the main thing we had to figure out.
Awesome. So as agent-based systems expand, what new challenges do you anticipate emerging at the intersection of AI-driven actions, on-chain execution, and the need for verifiable cross-chain system reconstruction?
Yeah, so I very, very strongly believe that we are at the start of what's going to be an arms race when it comes to the adoption of autonomous capital or agentic asset management.
And there are like two aspects that are going to fuel that.
And one of them is just like basically a competition
for efficiency, similar to what we've seen
in the traditional markets, you know,
after the advent of like funds like Renaissance
that made high-frequency trading and algos very, very popular.
The margins become smaller and smaller and you need to optimize in terms of your alpha,
in terms of infrastructure, in terms of execution. So that's one thing we're going to be looking at
very, very closely in the future. The second thing that's going to be fueling that arms race is going to be, you know, the price that makes all of that worth it.
And now, you know, today, the users in DeFi that are using agents have a vast advantage over, you know, the old manual users.
As agents allow you to get that 24-hour risk management, pull out of markets that are getting dicey the first,
like the human operator might be sleeping
while your agent takes care of that.
In the future, everybody will be
on a level playing field mostly,
but the reason why it gets so interesting
to have these agents work for you is that they're going to get access to a much more varied table of yield sources.
Today, pretty much all the teams we are talking to, the agents would send your money to similar places like a Morpho Vault and an Aave lending market, et cetera. It's a very, very highly correlated stablecoin-driven,
stablecoin demand-driven yield market.
But imagine in the future when you have RWA tokenization,
further along, you will have your mortgage yield on chain.
You could have tokenized compute.
You could even get yield on Bitcoin hash rate,
all these different things.
And then you would
have like agents that could curate counter cyclical yield products which i find super fascinating and
the challenge there is going to be it's not so much on our side but more so the the rwa the
a builders on this panel as well like like austin for example to make their RWA products composable and liquid, right?
Like if I have an agent that is constantly optimizing,
it wouldn't be able to do so much with an RWA token that has half a year lockup, right?
I would need a secondary market that allows the agent to get in and out of that market
quickly and efficiently.
So we would have to incentivize market makers to even make that market in the first place.
But I'm very, very excited about that future
where we'll have all these different assets on chain
and can really reap the benefits of that efficiency
of the automation going on.
Amazing. amazing uh so here is uh i'm gonna unless anybody has anything to
any any questions or anything to say on that
i'm going to go to the next one here to wormhole llama So Lama Risk aggregates signals from proof of reserve data,
settlement flows, macro indicators, and derivative markets, etc. How do you approach
such diverse data into a coherent on-chain net asset value safeguard?
Yeah, so I've talked a little bit about what we are building with LamaGuard NAV and that we're building it on CRE, but maybe I'll just go a little bit more in detail about the design that we're putting together.
Because I think it's a very simple, straightforward use case that serves a needed problem that needs to be solved.
So the problem starts with the fact that you have whitelisted assets
that don't have a secondary market.
And these have a NAV value that is delivered by the asset issuer themselves.
And typically, they work in conjunction with an independent fund administrator
doing accounting and delivering these NAV values,
which can be used on-chain as a price source for pricing these assets as collateral.
You can imagine that when you have a price source that is coming from a single entity,
even though they are regulated and they may have good, high-quality systems and have the best of
intentions, there are times when you may experience anomalies, which could be due to system failure
on their back end, querying the HTTP endpoint from their back end. Or it could be a legitimate solvency event that is affecting the fund.
So we need to be very careful about how we're integrating that data when we use it in DeFi
because we can't roll back the chain, and this can have severe consequences
if there's mispricing due to a system failure.
So we implement bounds logic that we run inside of the CRE workflow
that we are ensuring that the nav that gets relayed on chain
is within expected bounds according to the underlying assets
that make up that fund.
If it's a T-bill fund, we have certain expectations about its behavior.
If it's, for example,
there's a more exotic kind of collateral
that's listed on Aave Horizon.
It's Superstates USCC,
and this is primarily a crypto carry trade fund.
It exhibits higher than usual volatility,
and it has higher than usual volatility, and it has higher than usual yield compared to other more
traditional kinds of RWAs. And so we need to make decisions about how we expect the behavior of that
fund to evolve over time. And its behavior can change over time. So we need to have external signals that are helping us to make sure that our bounding logic is staying reliable to the expected behavior that a simple example is, is that the the federal funds rate has a direct impact on the pricing for for a T-bill fund. So we need to incorporate additional signals like that
in order to make sure that our bounding logic
has the most up-to-date information
and can deliver very tight pricing.
And the more tightly we can set the bounds of our pricing logic, that means we can
more confidently increase exposure of that asset into DeFi. So that means we can push the borrowing
power or the loan-to-values that are possible with that collateral and increase the total supply
that we're willing to have on exposure that we're willing to have to that collateral on the platform.
And these are important kinds of decisions that risk managers like LamaRisk ourselves,
we need to be making to give proper guidance to our partners like Aave Horizon
and how they think about exposure to assets and how they think about being able to be most competitive
against other DeFi marketplaces. So this is sort of the dual mandate that we have is we want to
make sure that our DeFi partners are doing everything that they can to be safe and robust,
in this example, using high quality risk managed nav data,
and also that they're implementing solutions in a way that they have the best shot at being able
to be most competitive and to most reliably attract users and show users like, hey, you're,
you know, you're not going to get wrongfully liquidated. Because of mispricing. We have these safeguards that are incorporated into our pricing methodology.
So you may have already answered this question then, but it was basically as things get a
little bit more complex, I mean, what do you see as the next frontier for cross-system
risk modeling?
I mean, especially as more asset flows begin relying on chain verification.
Yeah, so I talked about it a little bit with this example of Athena's USDE. I think that this is a
really great example. Athena's USDE has found a huge amount of adoption over this past year integrated into different DeFi strategies. And it's oftentimes
being used with pegged pricing, which we consider to be, as I mentioned before,
an intermediary solution before we are able to implement next generation, more sophisticated
risk managed pricings. And so in the case of USDE, it has some important features that we can create better risk oracles
to protect the DeFi protocols that are integrating an asset like USDE.
For example, it has an on-chain redemption buffer that Athena maintains a certain size
and replenishes it at an expected rate.
So we can use that as a signal that is if there seems to be stress with replenishing the redemption
buffer, then we use that as an indicator to potentially freeze markets that are using USDE. Or, USDE also has a proof of reserve that they update on a weekly basis.
We're one of the verifiers of that POR.
And we're working with them to increase the frequency that they deliver that POR
to more reliably be able to prove its solvency on a regular basis.
And so these kinds of risk signals we can incorporate into a pricing strategy that can more safely protect
both borrowers and lenders in these lending markets.
Awesome. Thank you.
And the last and final question here,
why do so many crypto folks identify as llamas?
So my experience with the llama meme is it seemed to start with the Curve ecosystem, as far as I know.
And I don't think it started for any particularly good reason, just that some people in the community had identified as llamas and then
everybody just decided collectively that llamas are the best animal so I stand by that stance
yeah that's that's great I guess uh I'm a little bit more popular than pandas then.
Yeah, yeah. All llamas here, no pandas.
I can kick people off this space.
Charles, I saw you putting the crying emoji. Was there something that you wanted to say?
No, no, no.
I just boomed everything up and didn't do the crying with laughter emoji.
Oh, yeah.
Yeah.
Oh, that's great.
That's great.
Well, so we're about to wrap up the space here I believe
I want to thank
everybody for joining the show
it was
a pleasure to host you all here
we're down to
our final
four guests Patrick
Charles Wormhole and Austin
I mean it's been an absolute pleasure
as we close out here,
I'll let you guys go ahead with a final statement for the show.
Actually, you know what?
Before I do that,
I have to shout out all the people in the audience.
So if you guys can do like a quick little 15 second last statement for me
so that I can shout out the audience, that would be awesome.
Awesome.
So Austin, if you want to start us out.
Oh, cool.
I just want to say thanks for everyone for listening.
Thanks to Chainlink Labs for organizing this.
It was great to hear what everyone else is working on with the CRE.
Awesome, thank you.
Wyrmho?
Yeah, thanks for having me
and great to have a conversation with you all here.
If you are interested in our work,
we have a public telegram channel at LamaRisk
and you can also check out our website, LamaRisk.com.
If you're interested in the work that we're doing with risk-managed price feeds
and development of POR using Chainlink CRE,
then feel free to reach out at either of those places.
Awesome. And I want to give a shout out down there to Luke.
Always a pleasure to have you in the audience and oftentimes as a speaker.
King Dane Kush, Victor Brown, the scientist.
Jack, who is the chief launch officer of LDA.
Shout out to you.
Jack may actually be a good person to introduce to Lama Risk, as my brain is ticking.
Nice.
Yeah, maybe we should
introduce them.
Based, KBot,
HHMC,
Shaman, Cool, Enrique,
Dirt,
NYCAN, Nick,
Greg,
Climb, George,
Zen,
Catechus, Rhythm, Diego, thank you all for coming here
Jalisco, Harrowing
Virto
Operations
Swaglord
Martin
Zidiziz
Awesome, thank you guys Patrick, do you want to give us
15 seconds? Sure, of course. First of all, thank you
Richard, Darren and Rob from Quickswap for hosting. Thank you to
all the other CRE
cohort members. It was a blast learning more about your projects.
Thank you to Chainlink and CRE for
being such great partners. And anybody here in the audience is building at the intersection of
DeFi and consumer apps in the neobank space, or if you're running a fund and you're interested
in making your cash positions more efficient and more fruitful,
then give me a follow, send me a DM and we can start chatting. Thank you.
Awesome. Thank you, Charles.
Yeah. I didn't think I could get more bullish on CRE,
but it's happened after this brief chat with everybody. Rock,
I really appreciated the questions. Absolutely brilliant.
Thank you very much for those. But yeah, it's really encouraging to see everybody taken to this.
And the fact that everyone's going to be using link tokens to pay for everything.
This is exactly what we've wanted for years since we saw the white paper.
So as far as I'm concerned, incredibly bullish going forward.
Thank you, everybody, for your time and questions.
Just out of curiosity, did the link ETF, is it still trying to go through
or did it pass?
Nice.
Nice.
Super bullish.
Thank you everybody for joining.
Sounds like a UK accent, Charles. So hopefully this was a good
excuse to get a
takeaway tonight.
It certainly is. I'll be getting something like some chips
or something. Maybe some chippy sauce if I can
wangle that. Yeah, I'm just a little bit
further down the road from you in Newcastle.
Alright, yeah.
Can you guys hear me? You've got decent chips.
Yeah, yeah, I can hear. Okay, I was
rugged for the last, I don't know, few
minutes. I wanted to just get up
and do some more old man
internet stuff. You guys were talking
about llamas.
So I think that, you know where it originally got popular?
It was either Minecraft or,
do you guys remember Winamp from like,
I'm pretty sure that was the 90s or late 90s,
when they had their phrase when you would turn on Winamp, which you used to have to use to like listen to music or any audio stuff.
and it would say,
Winamp,
it really whips the llama's ass.
Does anyone remember that?
Nobody knows the old Winamp?
Is everyone here like,
I'm fully,
I remember that.
Don't worry.
I remember that. Did they worry. I remember that.
Did they have color TVs back then?
I had a black and white TV at the time, actually,
but I think we were just poor.
Jeez.
So you got to see the automobile being built, right?
Okay, slow down there, caballero.
They were still using square wheels in rock state so only one person remembers
that when i'm thinking you guys go on youtube and type in winamp it really whips the llama's ass
it was like such an important part of childhood it was uh yeah it made winamp really cool
the early internet was a weird place.
It really was.
It was like crypto.
All the memes and NFTs and stuff and all the weird culture stuff.
The early internet was such a fun place.
It was...
All the coders would put in weird little
Easter eggs into the code
and weird stuff.
It was fun.
All right.
Now they're just
collected. and weird stuff is fun. All right. Now they're just exploited.
Now they're just collecting our data.
Yeah, we're just getting
harvested left and right.
It's all right.
And we do it.
And we welcome it.
Feed me some more ads so I can get my free
Facebook and Google.
Accept all. Accept all.
This was a really fun space. So guys learned a lot about Chainlink and CRE and all the stuff
everyone on the stage is building using these technologies. Really cool stuff. It makes me excited. This space, when we have more technical
and I really like infra and seeing all that,
not just infra, but the projects building the infra,
it always makes me really excited about our industry.
And when times are harder in the industry
and it's a little slower
and maybe we didn't have as much of a bull run
as we were all hoping for,
but I think there's still hope for it in 2026.
I actually think 2026 is going to be a great year.
Liquidity is opening up, money printers are turning on.
But besides that, I think it's really cool to see all the stuff being built in our industry.
And eventually, as Andreas Antonopoulos would say, we will have an infrastructure inversion.
as Andreas Antonopoulos would say,
we will have an infrastructure inversion.
All of the things that,
the rails that we're building on,
using early internet or previous web two
and all the other financial rails,
it'll all flip and they'll be built on top of us.
Just like AOL and dial up,
phones used to be on dial up
or run on the internet.
And now the internet,
sorry, no, internet used to run on dial-up or run on the internet, and now the internet... Sorry, no.
Internet used to run on top of phones, phone lines,
and now phone lines run on top of the internet.
And we're going to see that same infrastructure inversion with all these things like Siri.
All right, I am going to be late
and probably not make my flight.
So cheers, guys.
See you next time.
Bye-bye.
Bye-bye.
Later, everybody.
Bye.
Thank you. Bye-bye. Bye-bye. Later, everybody. Bye. Thank you.
Bye-bye.
Remember, follow everyone in his space except Luke.
Okay.
Bye-bye.
See you later, guys.
Bye-bye.
Bye-bye. Thank you.
Insights
Project L
P
