Nitesh Nath - DFlow: Internet-Native Market Structure - TOKEN2049 Dubai 2025

My name is Natashnath. I am the founder and CEO of Dflow. Uh today's talk is about internet native market structure. Um so kind of before I dive into the thesis and uh the talk will be slightly technical going over some market structural things uh in traditional financial markets as well as decentralized financial markets. Um before I get into that a little bit about Dflow.

So Dflow is a Salana Dex aggregator. Uh primarily we do two things. We build sort of this giant real-time graph of DEX's prices and liquidity on Salana. And two we spend a lot of time uh doing sort of optimization problems in you know low latency software calculating uh optimal trade routes between two tokens. Um this is slightly different from how things work in the Ethereum world where uh routing is done onchain typically.

Um we do all of our routing offchain and salon land uh and and then construct sort of like transactions that are then returned to wallets. So concretely for us uh as a dex aggregator we have two sets of sort of market participants in our system on on the taking side and the supply side. So supply side is uh liquidity from a lot of these sort of like dexes and um various onchain liquidity providers. Uh and then sort of the other side is is the wallets and wallet APIs. Um okay cool.

So let me let me talk to you a little bit more about what I like to call internet native market structure. Um uh this is you know one of the reasons why I got into crypto in the first place in 2021. uh my background is in high frequency trading. So was fairly well-versed with sort of this pattern that has emerged in uh in trading over the past three or four decades. Um a lot of trading used to take place in sort of like the pits of Chicago.

Uh and and what we've noticed is that every time there is a sort of like new substrate or new medium for price discovery to happen, market structure changes very very drastically. So, you know, decades ago when when trading happened in the pits, that was prior to the introduction of electronic order books, um, after trading moved to the electronic order books in the 2000s and 2010s, uh, lots of things changed. You know, there was the emergence of brand new types of of trading and trading firms. This is when Citadel Securities and the Jumps and DRWs of the world started becoming very big. Um in the 2010s till the sort of like end of the 20110s there was this massive latency race where all of these firms were trying to compete on sort of like landing their transaction or not transactions orders to to uh the matching engine which was a very centralized thing um in in sort of late 20s late 2010s and obviously you know Bitcoin came around during early 2010s but in in the late 2010s and what we have been seeing in the 2020s is the emergence of like a brand new type of substrate for uh price discovery.

And it's this incredible phenomenon that um buyers are meeting sellers permissionlessly on the blockchain. And uh I I think this is, you know, again, this is a phenomenon. It's a completely novel phenomenon, but I don't think we're going to start to see the long-term effects of of what happens uh over the course of the next several decades until probably uh later in the 2020s. So, this this talk is a little bit of history behind what has existed and um uh some predictions about what the future might look like. So um this is a very gross oversimplification of traditional financial market structure but the the purpose of laying it out sort of in the constellation of these three entities is to talk about how retail interacts with these markets.

So to date what we found in DeFi is that retail is the primary participant of these markets and a lot of these systems are designed on at least with what DFlow thinks about of protecting retail. So this is a very retail centric view on how trading happens with markets that retail has access to. So US equities typically uh in traditional markets. So uh the three the three entities that we have here uh are exchanges, makers and takers and brokerages. So exchanges are um uh centralized entities.

They hold matching engines. This is all stuff that you guys are probably familiar with. Um and they allow makers and takers. So the citadels, vertus, jumps, DRWs of the world to colllocate and trade on these uh venue, these centralized venues of price discovery. Now where retail comes into the picture is with uh brokerages.

So retail will typically use um consumerf facing applications built by uh companies like Robin Hood and Schwab. Um and uh retail will generate a lot of very soft order flow that uh the market makers love to have and the exchanges also love to see. Um so the the flow of the flow of sort of like order flow originates from brokerages uh is either internalized by the makers or you know directed to um uh third party like like the New York Stock Exchange various dark pools things like that. So in in DeFi and in particular on Solano which is uh you know where DFlow focuses and so what I think about the most you have a slightly different structure. Um so you you have uh first the aggregators which is uh what DFlow builds.

Um you have aggregators in traditional financial markets as well but they play a slightly different role. Um so you you have aggregators in in DeFi. They aggregate liquidity from AMMs uh which are liquidity providers. So um market makers effectively. Uh you have takers which are um me sophisticated actors and these these are uh you know typically actors that would like to take resting liquidity uh in order to profit pick off the makers and AMMs that are uh resting liquidity.

And then you have wallets. So wallets are again from the retail ccentric view of of how we like to view this market structure the primary place where retail interacts with the liquidity onchain. Uh wallets generate a lot of very soft order flow similar to the brokerages in traditional financial market structures. Um however the flow generally goes through aggregators and is executed by the AMMs. Um cool.

So the the thing that I want to highlight is um DEX is a misnomer in this sort of like new decentralized market structure. Uh DEX has the word exchange in it. So you might uh analogize it to um the New York Stock Exchange or uh NASDAQ or something. However, this is actually a misnomer. I would argue that a DEX is not actually an exchange.

They're just a market maker. Um and you know on Salana we're we're noticing a trend of uh AMMs that have become proprietary. So you know um uh typically the unis swaps of the world provide uh liquidity via capital that is deposited into the smart contracts natively on chain. Um on Salana, we're noticing this very interesting trend of proprietary AMMs that are run by sophisticated makers uh that are generally model driven, predictive of future prices and very sophisticated similar to the me sophisticated takers uh with um moving their quotes uh out of the way on on blockspace. Now the reason this is important is because I market structure on in DeFi today is very imbalanced towards takers.

This is inherent to the type of trading that market makers do and market takers do. So with market makers um they are resting their liquidity at all times uh on chain on Salana and market takers are opportunistic meaning they don't actually have to trade until they choose to. So with market makers having risk exposed and um uh sort of like DeFi blockchain market structure allowing uh validators and searchers to have sort of like privileged permissionless access to um market maker liquidity. This balances the favor um of this market structure very heavily towards takers. So what we are noticing with proprietary AMMs which don't allow capital deposited from retail is a a divergence from takers being the ones who the only ones who are sophisticated about accessing block space.

Um what does this actually mean? uh this is this is a very early sort of like predictive uh thesis that I have on on the future of Salana DeFi in general which is that once you start seeing sophisticated makers in a system this is a very early and strong signal of impending DeFi growth. Now the reason for that is because uh these prop AMMs in general are competing for order flow from uh these wallets and just generally any order flow on chain um in a way that is uh economically incentivized um meaning the the makers actually earn P&L uh rather than just charging a fee they capture the spread and then so they set the spread very dynamically and purposefully um and they they price the market in a way that is also sophisticated. So they're actually trying to uh quote around a fair price proactively rather than retroactively which is what like the passive AMMs like unis swap do. So um because you have this effect of of uh these prop A amms tightening to the the spreads in the market um uh trading becomes cheaper.

Uh every time a user wants to trade they pay less. Now on an individual sort of like trade by trade basis if you look at this sure they're saving one two five basis points not really meaningful when you look at it sort of like individually but aggregate this over 12 months and this is a lot of savings that um uh are either redirected back to uh users or um uh uh uh distributed back to protocols for for the use of better applications. So what happens is if if over the course of 12 months users are saving a lot of money when trading um you know they're going to have more money left to trade and you know the cost of trading if it's cheap then there's going to be increased trading. If there's increased trading market makers who earn P&L from basis points times volume if volume increases then there's more P&L for the market makers to have. There's more P&L for the market makers to have.

There's going to be increased competition. Increased competition leads to tighter spreads and the whole um cycle repeats. And there's a very strong byproduct of this DeFi growth loop which is that uh as more trading happens and more users arrive to the ecosystem. Um there's more protocol revenue and there's more application revenue uh rev is is sort of like one metric of this uh that is either reinvested into um protocol development or redistributed by protocols to token holders in in some form of the other. This is also sort of like this other uh growth loop that feeds into this the the previous cycle shown on the previous slide.

Um so reinvested, redistributed, increased user activity and then more apps and and more protocol revenue, more improvements to app UX and and more protocol revenue. Okay, so um what does this actually all mean? uh the you know I probably actually should have just put Salana DeFi will surpass Trafi which is the the main purpose of this talk is to highlight the fact that a a lot of the reasons many people entered crypto in the first place was to compete with traditional financial markets build sort of this new um system on a on a blank slate and and I think that uh over the next couple years it may not be so obvious today uh we we are actually on track for that. I think DeFi will surpass Tradfi. Um it's it's this you know completely novel market structure where uh instead of submitting orders to a centralized matching engine um you know makers and takers are are deploying their own contracts on chain uh they're doing so in sort of like this tiering complete language which is the language that the the smart contracts are written in.

Um and uh the the barriers to entry are are very low. So competing for flow is is uh as permissionless as as doing anything else in DeFi. You you deploy a contract, you submit a bid offer and and if you are the tightest bid and tightest offer at any given time, you win some flow. Um right, so that's the headline. Uh Salana DeFi will surpass Trafi.

Uh and this is something that we are extremely focused on at DFlow. Um and uh yeah, keep an eye on us. We we launched at the end of quarter 4, had a great quarter 1, integrated with Phantom, Jupiter's frontends, Camino swap, uh major applications, and and a lot more to be announced uh next quarter. Thank you.