Nikolay Avramov on the evolution of a top cross-chain DEX from idea to a TVL of $600 million "Unlike other solutions, we are integration-ready with any protocols"
3 August 2023

About the background

During my time at business school, I wrote a paper on blockchain — that was in 2016. By then, I was already working for a company dealing with traffic on Facebook, Google, and so on. In 2016-2017, a lot of projects were launching ICOs with just a white paper and a landing page, hastily put together in a few days. Various projects started approaching our agency for advertising and marketing services. By the way, I never defended my thesis because I dove headfirst into crypto — I was a marketer, a community manager in a Telegram chat, and worked on several projects that were launching ICOs. I then spent some time working as an analyst/marketer in a venture fund — our team launched a product that still exists today, a fully licensed Swiss crypto bank. Getting the license was one of the most challenging tasks, taking us about a year.
Later, I worked as a marketing director for a crypto exchange, and then, almost three years ago, I teamed up with a group of old friends to create what is now called Symbiosis.

About the project's start and its initial development

Initially, what we envisioned wasn't related to interoperability, token exchanges between networks, or liquidity movement. Our first idea was to try and combine the best of both the TradFi and DeFi worlds. That's how the name Symbiosis came about, from the word "symbiosis". It was coined by one of the partners who no longer has a corporate role or manages the project. What eventually materialized was quite different from our original concept.

Not every symbiosis of one thing with another is a good idea. I agree that we shouldn't replicate things from web2 into web3. It doesn't solve the problem. The one thing I'm grateful for from my business school education is the ingrained "problem-thinking" approach. My point is: creating a decentralized Uber or Facebook isn't a solution because there isn't a problem to begin with. Uber and Facebook work perfectly fine. But if there's a fundamental problem behind the idea, beyond mere replication, then it might be worth pursuing. The challenge lies in predicting unit economics, tokenomics, and other parameters.

It all started with creating another Telegram chat. Initially, there were 5 people. Now, that chat is essentially archived. We don't really have a headquarters, and most of the team is scattered across various locations, from the USA to Japan. We never had an office. However, we occasionally met in groups at certain places. In the beginning, our gathering point was Sochi. We registered in Seychelles because it's a relatively simple procedure there. But now we're looking at Asian platforms, as we need to consider regulation — it will inevitably come to DeFi.
In 2020, the BNB Chain was launched, and we saw a surge of replicas, lending platforms, NFT marketplaces, DEXs — all on the BNB Chain. By then, it was clear that users were interested in the market, Ethereum fees were unlikely to be low, and the emergence of new networks would continue, leading to competition. Yet, in 2020, there wasn't a straightforward solution for cross-chain exchanges. There were bridges, but a bridge in the traditional sense takes a token from one network and wraps it on another.

The issue is that the solution is makeshift. This wrapped token isn't traded anywhere and isn't recognized by wallets as a valid unit. Let's say a user wants to exchange USDT on the BNB Chain for USDT on Polygon. The challenge is to ensure that the exchange doesn't just wrap the token but integrates it into the existing ecosystem of DEXs, trading, and lending. This hasn't been done before. We spent about six months studying approaches to the problem. We conceptualized how it should look architecturally. Of course, things evolved over time — the second version is quite different from the first. We began coding in February 2021, with the help of a partner experienced in outsourcing high-load solutions.

In March, we participated in a hackathon by Chainlink. An essential component in creating a solution like Symbiosis is the oracle system that will transmit price data. Initially, we wanted to build our solution around Chainlink's architecture, but from a security standpoint, it didn't seem like the best fit. We didn't win the hackathon, but we drew the right conclusions and started moving towards custom software.
We used the ve(3,3) tokenomics because it seemed like the most suitable system. But we're currently revisiting it — liquidity providers will receive full rewards in stablecoins, while holders will get SIS tokens. We understand that we won't be competitors to Curve, but we definitely need to move away from incentivizing users with our token.
Rewards won't be canceled for those who staked their tokens. On the contrary, there will be more rewards for them. Currently, part of the rewards for those who deposit tokens into pools is given in SIS tokens — this policy is what we'll be revisiting.

About the pivotal moment in Symbiosis's history and new achievements

One such period was the significant liquidity outflow after the situation with Terraform Labs. At that point, liquidity dropped to 1-2 million. With such a scenario, swaps of amounts over $10,000 began to experience slippage, which was unappealing to users.
It's the chicken and egg dilemma. On one hand, you need to attract liquidity for something, but on the other, in such a declining market, it's risky to give out a large number of tokens. As the token's price drops, you end up giving out even more tokens, which people then receive and start to dump — it's a death spiral for the token. Moreover, without attracting liquidity, swaps simply won't go through.
Another moment was our launch in March; by April and May, we had few users. We decided to turn to B2B, actively promoting our solution and seeking new partnerships. We wanted to integrate into wallets, but this idea turned out to be a misstep because most wallets are protective of products like Symbiosis. Integrating into a browser is fine, but native integration of swaps, like Trust Wallet's partnership with 1inch, didn't work out.
However, our strategy with aggregators did work. The first volume we started seeing came through our partners, like Via Protocol and Rubic. But now, surprisingly, 90% of transactions go directly through our app, not through partner protocols.
Currently, Symbiosis's top four networks by volume are zkSync, Optimism, Arbitrum, and BNB Chain. Close to this top tier is Polygon. zkSync leads by a significant margin because we timely tapped into the hype associated with them. Plus, after Multichain ceased to exist, we captured a lot of traffic. Among the new networks, we'll first add Base and later Zora.
Why don't we support Fantom? We couldn't agree on marketing and liquidity provision. We've been in talks with them for quite a while, almost a year ago. It just didn't click. We removed them from our roadmap and still don't plan to revisit it because there's much more hype to be captured from various L2 networks. I actively follow Fantom, but I haven't seen anything innovative in their roadmap that could significantly change its positioning. Of course, a lot of liquidity is locked there, and many trades are happening, but I personally don't see any business there. We want to move away from the "add just for the sake of adding" approach.
Our TVL represents the amount locked in stablecoin pools. Initially, the TVL was 20 million, but it plummeted after Terraform's fall. The second blow came after FTX's decline, and although it didn't directly affect us, market sentiment and retail user behavior played a role. Currently, most of our TVL is backed by retail volumes and agreements with market makers. Over the past few months, our volumes have grown by 300%, exceeding 600 million in cross-chain swaps, and we aim to reach a billion.

About interest in other projects

Symbiosis doesn't compete with Uniswap X, and it's unlikely we'll act as fillers for them. Of course, we communicate with almost all teams working on interoperability, but we're somewhat different from Uniswap X.
Our plans focus on deepening interoperability within various subnets. For instance, Avalanche has several subnets and a high demand for quick liquidity transfers between them. We aim to assist with that. Ideally, we'd like to see such a situation in other L2 networks for Ethereum. The question is what additional use cases will emerge on top of that. There are definitely applications for rollups beyond strictly financial services. This is somehow related to scoring, gaming, and emerging social projects. It's very intriguing.
Layer Zero would have been even more interesting if not for Chainlink. Within its framework, you can create applied solutions like Stargate. Layer Zero has garnered significant support, but the question is how long their funds will last and whether they'll make it to the next investment rounds.

About seeking investors and promoting Symbiosis

Around July 2021, we began to attract our initial funds. It took us three months to close the first round. Two companies led the round — Blockchain.com and Binance Labs. Our approach to investors was straightforward — we started with cold outreach and searching.

It's crucial at this stage, when you hear the first "no," to ask investors to introduce you to other investors, VCs, and funds and to help with contacts. That's how it worked for us. Our first investment came from Blockchain.com, and after that, everything became faster and easier. For many investors, it's essential to see at least one lead. Then there's less focus on tokenomics and other parameters, and more on product positioning — whether it envisions the fund's portfolio. Not all funds like investing in competing projects — it's not always beneficial.

After Blockchain.com, we engaged with Shima Capital — from whom we received a significant allocation. It's evident that having good relationships with exchanges in the early stages is crucial because sooner or later, you'll need to discuss listings with them. We had our sights set on KuCoin and Gate. In the end, partnerships materialized with Gate and OKX, but things didn't work out with KuCoin. The reason was perhaps more personal — they invested an n-amount of money in us, but then asked for three times more for a token listing. We couldn't understand why.
Currently, we're running numerous campaigns targeting users from Vietnam, Thailand, Japan, Singapore, Hong Kong, and Indonesia. Based on our website traffic dynamics, we see that these users form our primary base, accounting for about 75% of our audience. Of course, we haven't entirely abandoned users from Europe, but our efforts in Asia are proving much more successful.

About partnerships and grants from Symbiosis

Currently, we're integrated into almost all aggregators, except for two — LI.FI and Socket. We plan to discuss this matter in a face-to-face meeting in Singapore. Otherwise, our main partners are those through whom we process all transactions on-chain.
When we planned to launch on Arbitrum One and Arbitrum Nova, we partnered with Arbswap. Now, we hold a priority position in their bridge section, which brings us high traffic. Our strategy revolves around creating a network effect with the integration of each new network. We integrate a network, automatically become the default bridge in the DEXes on that network, and then negotiate terms separately. We then position ourselves in all wallets and their dApps as the primary bridge. Interestingly, we've grown accustomed to being called a bridge. Initially, we resisted, emphasizing that we're a full-fledged cross-chain DEX, but now it seems easier for users if we just say "bridge".
We have a grant program — we issue grants to third-party programs for developing solutions on top of Symbiosis. For instance, Transfer uses Symbiosis for swaps under the hood. This service integrates as a payment system, for example, with a merchant — allowing purchases of socks with crypto. Consequently, users can pay with any token since the conversion happens automatically through Symbiosis. Last year, we issued 5-6 grants. In reality, we're open to granting even more, but we haven't yet prioritized this direction.

About the advantages of the DEX and its unique features

Firstly, we fundamentally allow the connection of diverse blockchain networks. I can't pinpoint other projects, aside from aggregators built on solutions like Symbiosis, that conduct semi-centralized swaps.
Via and LI.FI don't physically execute swaps themselves; they merely utilize one of our solutions and swap in a semi-centralized manner. We can physically exchange any token from network A for any token from network B, do it affordably and relatively quickly, and most importantly, in a fully decentralized manner. Neither the protocol team nor anyone else has access to the tokens during this process. This is all safeguarded by threshold signatures and a technology where the private key is divided into several parts, and no one physically possesses it at any given time.
The second aspect is that, unlike other solutions, we are integration-ready with any protocol. We can be used for cross-chain trading, NFT exchanges between different networks, and more. Additional advantages include the number of pairs we offer and our pricing. Plus, our unique feature — the octa pool. Essentially, it's a pool composed of assets with similar nominal values.

Of course, to ensure security when adding a new EVM-incompatible network, an audit is required. From the audits we've undergone, we can connect to blockchains like Near, Sui, Sei, and Aptos. But we haven't seen much point in doing so yet, as there isn't much transactional activity there. Cosmos is a separate issue, but historically and ideologically, we decided not to go there. And to say that we can start supporting Cardano, which has formally introduced smart contracts, at any time would be untrue because we're not yet equipped to do so.

Currently, our solution operates like minting and burning tokens. Token locking, minting, and burning are not mutually exclusive processes; they complement each other — ensuring security and preventing liquidity loss. All these methods are utilized in Symbiosis.

About cross-chain bridges without stablecoin dependency

Creating a cross-chain bridge that doesn't use stablecoins for gas payment isn't challenging; the real issue lies elsewhere. How do you determine the final price for the user? With stablecoins, it's relatively straightforward.
The price at a given moment, if your transaction isn't endless, is locked in the form of stablecoins, which you then exchange for the destination token and transfer to the user. This ensures a favorable exchange rate for users with minimal slippage.
The challenge is more about offering a user-friendly price and convenience rather than technical complexities. In the latest version of our octa pool, we've added the ability to include assets with the same nominal value, in addition to stablecoins. You can add ETH to the pool, which will manage ETH liquidity between different networks. It will shuttle it between networks like Polygon, Avalanche, Optimism, etc. Routing through this pool is possible.

For instance, users often need gas tokens for payments. If your routing only goes through stablecoins, and a user needs a gas token, you're taking several unnecessary steps. First, you take tokens from the user, exchange them for stablecoins on that network, swap them between networks, get the stablecoin on the target network, and then exchange it for a gas token. If you take a token from a user and exchange it for, say, an ETH gas token on Optimism, and the user needs ETH on Ethereum to pay for gas, one step disappears. Meaning, you're not swapping stablecoins with each other; you're swapping one ETH from one network for ETH on another and sending it to the user.

Thus, you eliminate one action, one gas fee, and the potential for slippage — making the transaction cheaper. We're currently researching whether there's a need to deploy other pools with different gas tokens, as ETH is the gas token in most L2 networks.

About ZK and enhancing security of bridges

Currently, our octa pool operates on top of L2 solutions for BNB Chain — that's just how it historically evolved. We're looking at ZK not just from the perspective of integrating different ZK networks, although we've already integrated networks like zkSync and Manta, but also in terms of using the infrastructure to secure octa pools.
We're considering several scenarios for the upcoming V3 update, where a ZK-rollup will likely be the controlling network for liquidity pools. One thing I can say for sure — the future is with ZK. I'm convinced that optimistic rollups will cease to exist unless they come up with something innovative or at least reformat into ZK-rollups.

About the development of zaps

Initially, it seemed like a killer feature that could disrupt the market. But now, I feel that the interest rates in various protocols aren't enticing enough to make ZAP a compelling product. Undoubtedly, had we launched ZAP before the fall of Terraforms, it would have been intriguing.
Currently, it's somewhat of a dead-end product. Perhaps we should not only allow users to deposit liquidity with any token but also assist them in rebalancing their portfolios. For instance, there are pools in USDT and USDC with varying rates, and users, depending on gas prices, can rebalance their portfolios to maximize returns. Such scenarios are appealing, and we're heading in that direction. We even have a prototype, though it's unclear how the market will receive it. For now, we're showcasing it to partners as a demo and gathering feedback on potential improvements.
In short, we have plans for ZAP 2.0, and work is underway, but we'll be able to present something a bit later.

About the potential of the TON ecosystem

Primarily, an ecosystem should be profit-oriented. Otherwise, it will quickly fade. You provide users with efficient products and solutions, and in doing so, you earn.

After all, we're here not for charity but to change the status quo, to provide tools to address users' future problems. But ultimately, people don't need anything — they're unaware of their current needs and the needs they'll have. Before the invention of airplanes, people weren't overly concerned about their absence and traveled in other ways.

I believe TON has a strong chance of breaking into the top 10 by market capitalization. I hope it does because there are projects currently there that shouldn't be, and I'm puzzled as to how they've managed to secure those spots. By the way, we've been wanting to integrate TON since the beginning of the year. But now, the ball is in their court because technically, we're ready to add support as early as next week. The main issue is the verification of our contracts. Personally, I think TON's products are useful and in demand.
Flirting with Telegram and NFT numbers might benefit the ecosystem's growth. The key is to keep innovating. Like merging Telegram's web2 solutions with blockchain. The burning question is, do we need blockchain here? For now, it seems so. The idea of marrying a financial service (payments, transfers, accounting, introducing bookkeeping) with a messenger where people consume and share information is the future.

About the battle between L2 and the potential victory of zkSync

In the long run, ZK will triumph — hence, zkSync stands a better chance. For now, Arbitrum and Optimism have gained significant traction and TVL because they're cheaper than Ethereum. Honestly, I think we don't need such a vast array of L2 solutions. Most will likely fade away over time, even though almost all L2 solutions raised massive funds at the outset. They'll eventually consolidate.

It'll be like the 90s when a plethora of browsers emerged. The most famous was Mosaic, then Microsoft came along and outcompeted the rest. People often prefer using 1-3 services rather than a multitude. So, the biggest ones will remain, and the rest will either cease to exist or become highly specialized.

From the L1 — Sei, for instance, declared from the get-go that they're creating a blockchain for financial settlements. And if we consider permissioned blockchains — like R3, they're essentially replacing the banking infrastructure, databases that are quite legacy and can't adapt to new financial products, open banking, open APIs, and so on. L2 might follow suit. But to say there will be a plethora of general-purpose L2s for everything — gaming, NFTs, etc. — is unlikely.
I'm unsure about the technical complexities of migrating from an optimistic rollup to ZK, especially when a project has significant TVL — it seems like a non-trivial task. It's like Ethereum's transition from Proof-of-Work to Proof-of-Stake. Whether the infrastructure and security model allow for a straightforward shift to ZK-rollups, I don't know. But I think if it were that simple, projects like Linea wouldn't have emerged.

About true decentralization and its universality

Decentralization isn't inherently good or bad. To some extent, it's simply an approach and a tool. A lens through which we view the world and interact with it. Like any tool, decentralization comes with a cost that we must inevitably bear.
Decentralization is great until you lose your private key. On one hand, it's fantastic because there are no censorship concerns. On the other, can Ethereum truly be considered decentralized if it has a limited number of major stakeholders? It's quite censorable. It depends on how you view centralization and decentralization.
Moreover, consider Multichain — a decentralized protocol, yet the private keys were held by the founders. As soon as those keys were lost, decentralization ended. Another example is any DEX. Everything seems great, but if at some point retail users withdraw all the liquidity, the protocol will cease to function because swaps physically can't occur. Therefore, a DEX needs to have its own reserve of funds, stored centrally on an NPC or some multisig, to replenish the DEX if needed. Technologically, this is decentralization, but ideologically, it's centralization. Otherwise, it simply won't work.

Living in Russia, the States, Europe, Ukraine — we don't experience the challenges of opening a bank account, but people in African countries do. They lack access to services we take for granted — loans or mortgages. Can crypto solve these problems? Yes. Of course, opening a bank account and loans are tied to personal identification, but combined with other technologies, crypto can provide a minimum solution. Will this create new problems? It will. But that's the cyclical development of technologies — some technologies solve old problems but create new ones.

This again resonates with the idea that we need to discern where blockchain is applicable and where it isn't. We need to address the current issues in web3 — and there are many of them. If Symbiosis didn't exist, I'm not sure what I'd be working on. Symbiosis emerged when Uniswap and various networks began to compete with Ethereum. Without the initial DEX model with liquidity pools and an automated market maker, the concept of a cross-chain DEX wouldn't have arisen.

About the development plans for Symbiosis

We're considering launching our own wallet and mobile app, but we're still assessing the necessity. We don't want to release just another wallet. We're actively exploring account abstraction with seamless stablecoin exchanges under the hood. We also aim to automate portfolio management using ZAPs — after all, we don't want to go all out and end up saying we've just created another trading bot.
We're looking into merging NFTs that we minted during marketing campaigns. There are several scenarios, and we're still figuring out the implementation. Currently, we have some applications for certain NFTs, but not those minted on marketplaces as images. We utilize the NFT standard 1155 technically, but it's for liquidity providers. Again, there are several scenarios, and we're still figuring out the implementation.
We've been in talks with Coinbase and Binance regarding the listing of SIS. With Coinbase, it's more challenging since it's a regulated company registered in the US. With Binance, we plan to partner and list the token as soon as possible — they've set certain conditions regarding user count, trading volume, etc. We're working to meet these conditions and revisit the discussion.

Perhaps we'll be the first to implement Vitalik Buterin's idea about smart wallets and bridge privacy as priority transitions. How this idea will impact the current bridge architecture, I don't know, but it won't directly affect us because people continue to use swaps. The proposal itself is intriguing. However, I don't see its implementation here and now; it will undoubtedly take some time.

We're contemplating the Symbiosis Chain — it will definitely be ZK-rollups. I don't think it will be our own app chain or blockchain that extends far beyond our use cases. To draw a metaphor — it'll be like dydx, which initially used Starknet — meaning only for settlements, to have a clear understanding of security and scalability of transactions within Symbiosis itself. If we provide a network and position ourselves as competitors to L2 networks, then it certainly won't be our business.
Article image
Bootcamp: DeFi Analyst
From a Reddit Gegen to a Dune guru.
Learn more


What is the usual learning process?
Will I be able to combine studying with working?
How much time should I devote to studying?
How long does the learning process last?
What happens when I finish the bootcamp?
What is the class schedule? What happens if I miss a stream?
I'm a complete beginner, can I study with you?
I have been writing smart contracts in Solidity for over a year now, will it be useful for me to study with you?
What is the difference between bootcamps? Which program should I choose?
What projects will I be working on? And what will I be able to do upon completing my studies?
Will I be able to get a job at a top-tier protocol in 6 months? And how do you assist with this?
I'm interested in crypto and Web3, but I don't want to become a developer.
Balaji was right