Fireside Chat With Bogdan Habic and Kartik Talwar

Right, we're going to jump right into our next session, which is also a back-to-back fireside talk. And for this one, I'd like to bring on Bton, who's the co-founder and CEO of Tenderly. Please give him a big round of applause.

CTO, CTO, CTO, and CEO.

Welcome. [clears throat] How has stick weekend for you?

Uh my voice is giving up slowly but surely.

Yes, you sound different already.

Uh well, yesterday was an interesting day. We saw the A V4 test net announcement. Uh but uh but a part that was hidden in there was the fact that the A V4 test net is actually 10 release virtual test nets. Um tell us about what they are and uh how that works and sort of what made how you kind of picked this.

Yeah. Did anybody try V4? maybe that they released no one ah good one hand thank you

two hands there [laughter]

so the idea was um a is releasing new version of the protocol version 4 and then uh the usual problem DAS and protocols have is how do we showcase this to the community because you can either wait to launch it at some point and then you're getting feedback but also you're kind of running a public company with the token and everything else so you actually want to protect yourself and showcase to the users and then hopefully with the bug bounty program and everything else for people to catch stuff on a test net versus on on a real network to for it to get hacked. And then we started working with Josh uh and the team on using a virtual test product which is the ability to fork a production chain. So they pro forked Ethereum mainet base and arbitum if I'm correct and then we connected them with the virtual CCIP. So also when you're doing crosschain transactions it's completely virtualized. So now for the next I don't know how many months or when they're releasing V4 exactly uh they can push out push out updates constantly to the A test net how they're calling it but it's actually a multi-chain test net and people can test it out with their production wallet only in a completely safe environment and I think it's super cool because this is what we built initially virtual test nets for and literally no one used them uh everybody used them for CACD which was one of the other use cases but we thought this was the cool one and I'm very thankful for to Josh that that really gave us a chance to showcase this and yeah people really like it so far.

So was it very much of a scenario where people kind of discover this is like a really good way to kind of reduce the feedback loop or do you feel like it's finally working because now iteration is kind of the mindset of a lot of the companies here and they kind of need to do this or or they have no choice to raise everything once. So I can tell you how it started a conversation with Josh. Uh bridging between two sapoleas. I want to say ETH mainet and base. I'm not sure.

Yeah.

Do you know how much time it takes to bridge transaction between the two test nets

for base versus ei?

So to use CCIP between baseia and main and sepia. I

say 1 and a half hours.

No, it's 12 minutes. But still now imagine while you're at a developing all of this and you're a UI engineer or whatever or you have a test running and you and you click bridge and you wait

for like 50 confirmations 10 minutes.

Yeah. Yeah. Yeah. Yeah. So it you wait 12 minutes and it's like make a coffee type of situation.

But there's some other chain pairs where you need to wait 12 hours. That's not even like you have a day

and then

my code's compiling and I'll come back tomorrow.

Yeah. Tomorrow.

Yeah. It's like from all those old IBM days where how I was called at least they like you write it and then you set it foration started. No, but um this is how it started. So much than what you said. Uh but no, I I think this is important because normal tests have their place as well.

I think they're very important for node operators to validate hard forks or for validators to or stakers to test out all of the all of the upgrades that happen there. I mean, we're a huge user of normal test net because we run so many nodes. But for if you're an application developer, protocol developer, uh if you're an institution on boarding node, onchain or etc. Um you don't need a normal test net like you can either develop locally or if you need a collaborative environment either within your team or your or your

but like locally you don't get to access different states because you can't for necessarily a version of kind of history stuff that would be the biggest selling point.

Yeah. Yeah. Technically because you're reusing reusing the main and state. So you don't have to take care of your test net where all of the teams stop. And then with virtual test that's another thing that I'm very proud how the team made uh we can sync the state.

So whenever for example all the price fees are updated they get updated on across all of the staging environments in the space that use this as well. So it's constantly live with all of the production data but you're safe.

Gotcha. Um on kind of similar topic but changing themes a little bit. Um, tenly honestly has been in the space for many years, but more than five now.

Eight.

Eight. Oh, there you go. Like, I'm pretty sure it's definitely before.

Our new head of customer success told me, I have to say this in every conversation that we're uh 8 years old, 90 employees in the series B company. And I'm like, those are vanity metrics, but I'm an engineer by you know, I remember you from in Paris from 2018. So, it's definitely seven years if not more. Um the question I have is like obviously a big part of what the early years have been were around crypto companies trying this and just trying to get a better developer environment but I assume you're getting a lot more customers of interest rather than non kind of crypto companies where whether they're financial institutions from people trying this thing enterprise software um how does that influence how you think about building better developer tools and what are they asking for are there kind of obvious things that are differences between those two groups

I'm I'm guessing here you're referring to the I mean everybody's mentioning the art 10 those and like

ML2s or even just like banks saying you want to do something internally whether it's resettlement but we need ax for this type of developer environment to be true for us because that's what our expectations

so the interesting there's two things I can say to that so there's less of an influx than people would expect so it's not like we're talking to 50 banks or whatever now also I think I don't know Sundep they have very good BD teams the chainlink crew uh a morph my guess is they are exposed to much more conversations but they think tenderly is a very good proxy of how many engineering teams are thinking about stuff so we're a very bad proxy how many financial teams think about stuff but for engineers uh I honestly think even though my job as a founder is to say that really cures cancer I think we're the de facto destination for for developer tooling in infra um what we see is um the teams are much smaller than we would have liked. What do I mean by this is I really hope that like City Bank came to us and they're like, "We need 1,500 seats for our engineering team that is implementing 700 stable coins, 40 networks, and we're going to give you a million dollars to give us all the infra." Well, in reality, it's like, I have these five quants and three solidity engineers. Poke them with the stick until they do.

And we sign up for the free trial account [laughter]

in seven months of procurement. But no um this is too true of the story until goes too far. But um what we're seeing now definitely is that the engineering and from what we learned quant teams so the people in the digital asset vertical we can call it that way definitely are exploring uh stuff on chain. We have some that are more far ahead. So the fintex are definitely more far ahead with like the revolts of the world stripe with tempo circle with dark arc um etc.

But we also see traditional players either dabbling toes and then there's some who are fairly advanced actually in their in their thinking which is which is very interesting.

How does that thinking translate into needs? Does that mean like if they're advanced which means we're translating that to they are really well with the max they can get it which means like you already support these things you have to really satisfy any additional need or does that mean they ask for more than defi teams? Yeah, I think the I mean the main difference is when we were talking with them it was interesting. I can't name the the company but I can say that when you're

What does it rhyme with?

What does it what does it rhyme with?

Yeah, it would be funny. Di no it's like no we haven't talked with city but um no it's interesting because like you cannot share your screen on a call with them. So I was in a situation where I held one laptop in front of another laptop to show a demo because you cannot share a screen to the to the engineering team inside of the institution and uh you can only send them PDFs, you cannot send them links to it's like a bunch of these weird things. So one stuff that's definitely happening for developer tooling which I don't think we should lose too much time about talking about but it's like the all of the compliance stuff that exists that exists there which my brain still doesn't comprehend but we're going to figure it out. Uh the other thing is um I I think the for them it's much more how do we integrate stuff than how do we build stuff and I think that's where the where the biggest different lies because I don't believe I I might be wrong by the way completely none of this is financial advice as they say but I believe that these banks will be wrapping morphos ais of the world they will be deploying unis swaps or paras swaps or aerad drromes or whatever because now I think what's interesting is that all of these money Legos or financial instruments that we have been building.

These are now becoming developer tools for them because I I need borrowing and lending. So I get Stripe similar to how I know Stripe is a payment processor but then as engineer I use it as a developer tool. I need to accept payments for my SAS products. Stripe is a developer tool for accepting payments. This is why I integrate Stripe.

Similarly they're like hey we need liquidity here to do something on top of it. We need per we need whatever. So what I'm going to get I'm going to get unis swap. going to get more for or whatever and I'm going to put it into my mobile app. So the differences here is and how we are looking at it now is tenderly so far has been focusing so much on helping people build onchain.

Now we're more and more focusing on how do people integrate onchain stuff and how to make sure that once things blow up uh they can show that they did everything possible in their power that to show that

they should kind of the answer to that question cuz does that mean you're kind of letting people deploy things out of the box and that's kind of the approach is that what the direction becomes or are you saying that uh you want to give more visibility into this but not as to help them achieve that.

Sure. So when you are a person at Swift again all of the examples by the way that I'm giving the ones that would be real I would be under NDA so I'm giving examples with companies that we really haven't talked with if you're at swift like in I know they had the CCIP thing all of the stuff uh all of the stuff that they do they necessarily don't have to write even a single line of solidity most of the value of Tenderly so far has been hey I'm writing solidity where did I make a mistake now for swift where we help for example in that case is hey I integrated I know most of the PR was like chain link runtime environment in CCIP and I'm doing it somewhere the value is hey all of these off-chain parts that I've used to do onchain transactions in a decentralized way uh if something goes wrong I like to joke who killed the butler so was it CCIP not being uh fast enough was it me preparing the transaction payload incorrectly which is different to me writing the piece to solidity code. Um, was the price speed updated? Was the node RPC up? Did the oracle return the information that I wanted?

Did the bridging happen when it should have? Uh, did it get the right amount of assets, etc.

Really great observability effectively.

Yes. I mean, the term that we really want to push for is chain operations. So, how you had developer operations or security operations or DevOps and secops, now it's like chain operations. So, chain ops, which is like how you have application performance metrics. Now you would have blockchain performance metrics because our stacks are fairly it's not even more complex as much as heterogeneous than than in in cloud because in cloud everything is cloud here it's like a combination of inevitably especially for these institutions going to be a combination of onrem which is like Fortnox level

and just cloud

and cloud yes and then web three and every the web three person wants to think of the other two as black boxes the fortnox person doesn't talk to the probably the cloud person is like I'm going to kill myself also be be black boxes. So here integration reasoning about stuff is key versus versus anything else because at the end of the day I think the incentives inside of the organizations were different. Here we were all building stuff for the sake of technologyization etc. Here they are seeing an opportunity to disrupt how currently the financial system works, which by the way they also run. So, but they're trying to figure out you have the top 10 financial players, whatever.

And they're trying to throw the board in the air to see how the pieces land now because this is this is the thing and then hopefully not get fired at the same time. So, that's the IBM

approach of we'll try it out and see what happens.

Yeah.

Um maybe let's zoom out a bit. Um the thing I guess I'm curious about from your perspective is understanding how do you think the directionality of everybody activating their own L2 now is that a world where that just seems to kind of be the reality because it's so easy now to kind of press a button and now build your network you know somebody has to support somebody has to figure out how communication or do you feel like there's consolidation or have people try it out but then they will kind of just realize how impractical or hard it is and you know change lines again. What do you kind of believe is the convergence?

The whenever I talk about L1's and L2s or uh how did we call them settlement and execution environments an hour ago. Uh I would like to define for L2s two different parts. One is the or how at least I talked about them ecosystem chains versus app chains. Now most of us would probably have a similar definition but in two sentences what I think about when they say that is ecosystem chains most of the block space is used by a lot of apps application chains most of the uh block space is used by a single app this is the

so so as an example base would be assistant chain but unis

yes or zora is was a better example

on top of base too

yes on top of base or an interesting argument if you would really like to catch me and my logic would be was optimism a world coin app chain when they were doing like 70% of of volume but

worldcoin is appchain now

yeah well now

but but before yes yes before it would be um now why do I mention this distinction is because I don't think that we will live in the world in a world where we're going to have 50 ecosystem chains like there isn't going to be 30 arbitrums 60 bases is 20 optimisms 100.

So this is a statement about postrooming like they might be 100 but not

yes

all will remain.

So the only way how I see that we have triple digits of these chains if is if uh countries start spinning up their L1's this is the only way or and L1's or sorry ecosystem chains or L2s that are ecosystem chains. This is the only way that they see generalized block space being deployed that way. my my um I would say we probably are off by like a few on that mainly because I can see the argument that maybe the top two or three banks in every country will also have their chain and you kind of just still create a similar network but

yeah it largely agreeable

but the question then is are those I'm going to touch briefly then afterwards on payment rails and and some other stuff as well but I think that we're going to be having thousands of app chains so like a video game having its own chain etc. I really still believe in in the future at least to rephrase the ecosystem piece the ecosystem chains as we see them today I don't think we're going to have more than like 30 40 I I don't know the number but I honestly think that's going to be the case

100%

now I agree with some deep where he's like of course Stripe launched an L1 this is like they're large enough they can do whatever they can coin a term L minus2 and launch an L minus 2 like it's it's whatever um technology Logically, we can have a debate if it makes sense. Business-wise, I think it makes complete sense because

Coinbase launched one.

Yes.

Seems to be working.

Yes. But they think it's a completely different uh marketing and positioning because

Oh, sure. But but but the argument was uh being able to actually create the demand for

Yes. Yes. Yes. Yeah. I agree.

Anyway, um one, two, three. Yeah. So, it made complete business sense for them to launch an L1 because they're they have more than enough money to push it onto us, if we would say it that way. Um, now the interesting thing is that for all of these chains, we see stable coin plus chain combo. So, Stripe bought bread.

Currently, we don't have a stripe stable. We're going to see if we're going to get a stripe stable. Uh, USDT has stable and plasma networks. Circle has USC and USDC and the ARC network. I know it's an interesting question.

If JP Morgan launches a stable coin chain, does it have a JP Morgan uh stable coin? I think that in the next year, two years, I don't know about the timeline, we are going to always see them deploy these things in pairs. So, both the network and the chain because the the downside is you spent a few millions, which is like literally a running error for these organizations. You spent a few millions of infrastructure and that's it. Maybe your stock goes down.

Hopefully your stock doesn't go down, but the upside is so uncapped. It's literally like the next version of the financial system. It's like dumb.

Yeah. Like what's the market gap of money?

Yeah. Money. Money. Yeah. Yeah.

It's It's literally It's literally that argument. What I think will happen afterwards is that we're going to see the consolidation of these like web 2 RAM L1s and L2s where we're going to start seeing stable coins move between them. Hopefully, if we have interp on them. Um the example that I have been giving is the joke I made I think three times this week on different panels but uh my Eastern European mind cannot comprehend the fact that Macy's has a credit card. So let's start with that or United or whatever.

Now in the post world of [snorts] all of this happening maybe it still looks like a credit card or whatever. If you think of where can Macy's or United have that level of integration they can have it at the wallet level. They could have it at the network level or they can have it at the stable coin level

and all and the app level

and the app level itself.

And the app level itself. Yeah. And an application.

That's a loyalty. Sorry. Yeah.

Now if we talk about wallets, I would argue that Door Dash, Uber Eats, Amazon are wallets of the web two world because you use something that has money to do something.

They give you credits too. Yeah.

Yes. Um, and then we are left with so will Macy's or or United create a wallet depending

actually United has a wallet cuz if I get a refund it goes to my United wallet for my flight.

Yes. I didn't know. Okay.

But but it's like a

I have the history of in my American United has a wallet. The rest of the world I don't know about

but imagine me switching. Actually in my app store there is no but anyway going back to this um my assumption how stuff is going to pan out is that United the long most logical way for United to do something here is to issue a stable coin where whenever you transact with that particular stable coin what do you get back you get points that you can use for miles because a wallet okay sure United is a great is maybe a large airliner and it makes sense but it doesn't make sense for them to run a network. Why would they? It doesn't make any sense for them to run a wallet. I really think because again, we have how many delivery apps?

It's like three or four per geography. So, it's definitely going to be some consolidation. So, but you transact a lot like you can use multiple credit cards, etc. So, stable coins makes sense for me at least.

Is there something you can do to prep for that world whether as a developer or just as a consumer? But does that mean like we kind of have to figure out smooth interoperability or you just kind of get up and effectively have them figure out what they want to do cuz you're going to have to have money move between the parties.

I think that so I don't think there was ever a future where this moment wasn't happening because crypto would have died like there at least I believe there was never a way to make a parallel system and just kill the old one. I I think that was highly childish to think and sorry if I'm attacking anyone's worldview and I'm on stage and they cannot shout back now. Um I think this moment was inevitable if we think that crypto will win. I think it's on us now to fight for decentralization, privacy, standardization, how no matter how boring it is and all of that stuff because this is our Trojan horse going into the door right now and we need to figure out what drops out of the horse and this is the important piece that needs to happen now. Be are your developer or a lawyer or or Sundep for that matter or whoever.

No, no, but this is really the case and it's all of us to help where we can. what we are going to help with is for their developers to see the benefits of web 3 and not just consider it another API call. This is what we can do and then hopefully some of those developers are going to be like hey actually this onrem thing that we're running and it has been seen the internet for the last time in like 1997 when we disconnected it from like in the basement maybe that can be a decentralized uh decentralized node somewhere as well and I think everybody needs to do their part there. Any uh parting advice on how founders should think about building in the space? Any parting advice on how founders should think about building in this space?

Because it seems like the world is pushing forward maps and revenue as the things that are interesting, which is how the rest of the noncrypto world already works. Um, but you kind of have a lot more choices and that usually leads to not the best outcomes because you don't get too focused and there's a lot more cost of just opportunity here, right? So, uh, what's kind of the the way people should think about building companies and uh what do you asked to attend think uh their time and attention should go to? I mean, we were the unfortunate I honestly think this is unfortunate statement that I'm going to make. We were the odd one out of not doing a token and being a business.

I don't say that token based businesses aren't businesses. But I'm just triggered when we say 2025 is the meta of revenue. All years ever have been the meta of revenue just to like we live in a capitalistic world. Anyway, uh so in that sense, I think now it's a very interesting time to raise money for a non-tokenbased company. I think that's one interesting thing.

I think even if you raise money with a token right now, now I think there won't be so much pressure for VCs. I might be completely wrong, but I don't think there's going to be so much pressure from VCs to pump the token price versus actually figuring out the business model because you run your company completely differently because one is like marketing announcements and our BD team is making whatever and the other one is like figure your stuff out. Um, and I think that now there's a lot of opportunity a lot of opportunity to be in the middleware. It's it's the boring piece of software, but like this is what makes generational companies like that middle layer that connects the two worlds like and we need more companies there. We need we don't need I know this is going to sound unpopular.

We don't need another vault project. We don't need another AMM. We we fixed that. That's done.

Do we need more stable coins?

We need thousands of more Mac stable [laughter] coins. Yes.

J, thank you so much for that amazing conversation. Please give him a big round of applause.