Listen to our podcast with Ember and learn how the biggest markets are efficiently priced, and the real alpha in DeFi hides in the dark corners — obscure chains, assets, and strategies that most allocators never look at.
In this episode, Nertila from Trading Strategy discusses with Joe from Ember Protocol. Ember is the investment platform and infrastructure for launching, accessing, and distributing traditional and onchain financial products through crypto capital markets.
- View Ember vaults and yields
- Watch the Ember episode on YouTube
- Listen to the Ember episode on Spotify
Any questions? Join our community Discord server.
Transcript
Nertila: Welcome to TradingStrategy.ai, the podcast where you can learn more about essentials of DeFi trading and new vaults. This is your go-to show for automated trading strategies and market insights. Welcome to the conversation, and this is your host, Nertila.
Nertila: Hello, everyone, and welcome to today's podcast of tradingstrategy.ai. Today, we have a new guest. His name is Joe, and he works in a project named Ember Protocol. Hello, Joe. Welcome to our podcast. Thank you so much for joining me today.
Joe: Hi Natila, thank you very much for having me. It's a pleasure to be on the podcast.
Nertila: I would like to know about your crypto journey. How did it start for you?
Joe: So back in around 2020 I was working in enterprise software as a project manager. at that time I was working on a large digital transformation project for the Ministry of Finance in Cyprus. I had a friend who started investing in crypto and he would come round to my house and try and shill me on the latest crypto that he was buying and show me how much money he was making.
Joe: And I'd been aware of Bitcoin for a long time, but I never invested in it. And I was always quite skeptical. So pretty much he would come round to my house and would try and argue why this is the future and I would argue why it's not. And I think I didn't fully understand it and neither did he.
Joe: But there was a certain point when he started to explain Ethereum and smart contracts and DeFi, and that began to really make sense to me because the software I was working on was enterprise resource planning software, which is essentially a big centralized ledger of transactions that operate according to some defined rule set.
Joe: And what he was describing is something quite similar, but it's a big decentralized ledger with autonomous rule enforcement and something clicked there. It immediately made a lot of sense and I could see that there are huge possibilities to this kind of technology. So I went completely 180 and flipped from being a skeptic to a fanatic and I started researching everything that I could, reading everything that I could investing and getting involved in DeFi as well.
Joe: I was very keen to- Try out anything that I was investing in firsthand to see if it was real or not. And so that's how I started getting involved in using DeFi. And whilst I was in Cyprus working there, I found out that they had this, at the university in Nicosia, their first blockchain and digital assets master's program, which I thought was interesting.
Joe: So I then spent the year studying at the university there. And that was quite interesting because we got into the technology side and the economic side, and also some of the sort of social and philosophical side of this technology. And following that, I then started working for Pangea.
Joe: At the time, the company was actually named Superchain, and I worked at Pangea for three years from 2023 to 2026. We built an incredibly high performance multi-chain indexer in Rust that could index over 500,000 transactions per second, could handle EVM chains and Sui chains. And then we also built a lot of AI functionality in that to allow people to explore blockchain data with AI.
Joe: And it was during that time that I also got very interested in data science into what's going on-chain because we had all of this vast access to on-chain data. And I was continually testing the platform and exploring the data, and that's what led me into building DeFi strategies and doing what I am doing today at Ember Protocol.
Nertila: in fact, I want to talk a little bit your current venture Ember Protocol. What is it exactly about, and what kind of blockchain layer are you using for Ember?
Joe: Ember Protocol is a vault platform. We build the infrastructure that allows people to operate DeFi and real world asset vaults.
Joe: In terms of what chains we are on Ember Protocol is really chain agnostic because vaults the strategies that funds are deployed into in theory could be on any chain, and in theory, deposits can come from any chain. Currently, our vault contracts are on Ethereum Mainnet and on Sui Meaning users can deposit assets into Ember from either of those chains.
Joe: But the actual strategy, it could be across multiple different chains or it could be off chain if it's a real world asset strategy. That is Ember Protocol as vault infrastructure, and we also have our own in-house curation team where we curate our own vaults, and my role is I am the DeFi lead at Ember, so I'm leading our curation team.
Joe: I'm researching and deploying new strategies, identifying new opportunities on the chain and also with potential off chain deals, and also giving feedback to the team on the product and our vault infrastructure
Nertila: Basically what is your day-to-day job with Ember Protocol?
Joe: I would say part of my day-to-day work involves research. So I'm continually looking across the different opportunities on chain to try and identify where the best risk return is currently and how best to allocate assets and manage the risk.
Joe: To do that, I use some tools that I built myself that basically help me scan across the whole chain a kind of universe of opportunities and give things a relative scoring to see where best to allocate. And I'm also looking for new strategies. So aside from allocating within the strategies that I've already set up we're looking at new assets new chains where we might deploy new types of strategies and I'm building testing frameworks for those to see whether these trades are going to be profitable for us, what the scalability of them is, and so on.
Joe: I'd say that's maybe a third of my time. There's also the day-to-day tasks of actually operating these strategies. Another part of my role is partnerships, so reaching out to other projects in DeFi that we might want to integrate with, and also speaking to other curators who might want to launch vaults with Ember, try and understand what their offering is, assess the quality of it Whether they are people that we would want to bring onto the platform and work with.
Joe: And then also a bit of business development to try and find new LPs for the fund, try and communicate what the risks are of our different products and basically to try and grow Ember to be one of the largest and most successful vaults platforms out there.
Nertila: You did mention that you focus a part of your time on risk management. I would like to know more what kind of system do you have in place, and is it only technical or it's also conceptual part of the project that you are running?
Joe: Yeah. So in terms of risk management systemI was working at Pangea where I was very closely involved with building a very comprehensive system for ingesting on-chain data and analyzing it.
Joe: So I have that background and experience in on-chain data science. And when I came to Ember and started to tackle this problem of vault curation, I was looking across the space at different off-the-shelf platforms available for portfolio and risk management, and I wasn't completely satisfied with them.
Joe: I felt like they will meet maybe 80% of my needs, but then the missing 20% is quite crucial, and they don't allow enough flexibility to customize exactly to what I need. So I ended up developing my own portfolio and risk management tooling. It's a system called Minerva and that allows for continual monitoring of positions also a back testing suite to try out different trades and see how they might perform and on-chain alerting and circuit breakers and so on.
Joe: So I think in general, this space has had a bit of a poor history of risk management, and I actually think it's one of the most crucial things that's going to differentiate which curators are going to succeed in the long run. We've seen many spectacular blowups of different vaults and different managed strategies, and many times they- Come down to poor risk management.
Joe: I think that this space unfortunately, it has been quite short-sighted in pursuing sort of high headline APY numbers without really considering what the underlying risk of those numbers are. And there's a general sort of misunderstanding that if one thing has 5% APY and another thing has 10% APY, then obviously the one with 10% is where you should put your money.
Joe: But that's often obscuring what the risk is behind those numbers. So my approach is really to try and deliver the best risk-adjusted returns and not simply take on undue risk in order to give higher numbers.
Nertila: So where do you think the capital and risk will accumulate in the future
Joe: so I think that there will be a bit of a split. I think what the market is really looking for is low risk yield primitives with senior protections that can serve as very high quality yield bearing collateral. So we just unpack that a little bit. We've seen the rise of loop r- looping and leverage products and now also risk tranching.
Joe: And these are allowing people to get higher returns by essentially taking on more risk through either leveraged looping or through being a junior capital and taking the first loss in the event that happens. And I think what the market is really looking for is an asset that is going to be relatively stable and bring a return that is somewhat above The risk free rate or the borrowing rate which allows them to then leverage that as capital.
Joe: I think there is less appetite for products with a higher risk profile or with more volatility precisely for this reason that they don't serve as quality capital and that then limits the ability to deeply integrate them into DeFi and use them as these composable money Legos. I think if something is going to be a good money Lego, then it needs to have a constrained and understandable risk profile that is somewhat predictable and isn't suddenly going to go to zero tomorrow.
Joe: But I also think that there will be a role for, in a portfolio, small allocations of risk capital with some manageable downside. And so I'm very keen on products that are allowing the market to properly hedge or structure their risk as desired.
Before we get back to the episode, if you are interested in getting access to over twenty-five thousand drywall datasets, including performance and risk metrics, visit our website tradingstrategy.ai for free data and professional subscriptions. Now, let's get back to the episode
Nertila: What are some of the hard lesson that you have learned so far?
Joe: good question. When I was early days getting into crypto and DeFi, I certainly made and lost large amounts of personal money from directional trading. Mostly just trading Bitcoin and Ethereum on leverage. And after many times going through the cycle of making and losing lots of money, I finally came to the conclusion that my personality was not that well-suited to directional trading.
Joe: I read a book that I would highly recommend called Market Wizards. There's also a part two called New Market Wizards, and it's a series of interviews with very successful traders from traditional finance. And my biggest takeaway from that book is how different lots of the individuals interviewed are in terms of their personality, their approach, how they were successful.
Joe: I thought it was very interesting that many of them had many failures had lost money, a lot of money repeatedly, but had come back and had eventually found a system or a way of working that fitted them. And so the lesson that I took away from that is- In order to be successful in trading or in asset management, you need to find a strategy or a way of working, an approach that fits your personality and who you are.
Joe: And as well as doing some directional trading, I was also getting into doing some arbitrage strategies and I found that they were the most consistently profitable and that it suited my way of working a lot more because it was something I could fully understand, I could map out I could optimize it and it brought very consistent returns.
Joe: So that was the hard lesson I learned is I'm not well-suited for directional trading, but for finding and exploiting inefficiencies in the market that's where I can really add value.
Nertila: So DeFi ecosystem has been plagued with lots of hacks lately. Where do we go from there as we have seen some of its impact on the market and also on the sentiment of traders?
Joe: Yeah, so I think there's two directions that we need to go in as an industry to help solve this problem. On the smart contract security side especially in the age of AI where if you have bugs in your smart contract, it's increasingly likely that they will be found and exploited and making your project closed source will not help anything.
Joe: There is no security through obscurity because AI can very effectively decompile bytecode now. So on the security front, I think maybe one of the only ways we can really be sure that something is secure is through formal verification. Formal verification is a technique of providing a formal logical proof that a system will behave in a particular way.
Joe: Ember's smart contracts have achieved a formal verification and I'm very keen on this approach for security. I know also Viper, which is a EVM smart contract language, has been making big steps forward there. They have formally verified the Viper compiler, and they're now also working to formally verify some DeFi primitives like the Curve StableSwap contract to show that there is logically no way that this can lose money.
Joe: I think that's very important for giving people certainty about security. Otherwise, there's a nagging feeling that, as we've seen with supposedly battle-tested protocols like Balancer, maybe there is some exploit there that no one's quite got to yet. So that's on the smart contract security side.
Joe: On the market side, what I think is really important is giving the market the right tools to structure risk correctly. And a few examples of this would be insurance. Nexus Mutual has been a leader in that space in providing smart contract insurance and paying out if hacks do occur.
Joe: There is a new project that we will be integrating with and announcing shortly, which is OpenCover, and they are offering insurance coverage for vaults, and essentially they stream off a small premium of the yield earned by the underlying vault in return for providing cover in the event of de-pegs or smart contract exploits or Oracle hacks or a range of different conditions.
Joe: So I think that's a financial product that allows the market to better manage their risk as desired. And another integration that we'll be launching shortly is with Reiko Dawn to have a senior and a junior tranche for our eEarn stablecoin vault. And essentially a senior tranche gives a certain level of downside protection regardless of what happens.
Joe: So whether the strategy loses money just in a normal way or it's due to some hack, some exploit in one of the underlying allocations, the depositors in a senior tranche would be protected as the junior tranche would take the first loss there. So I think allowing the market to correctly express its pricing of risk and giving it the tools to structure or transfer risk is also a very important part of the picture here.
Nertila: When I was trying to schedule this podcast with you, you did mention one of your favorite topic is the efficiency of the market. Can you di-dive deeper into this topic? I want to know more like how efficient is the market today?
Joe: Yeah. So one of my early obsessions in DeFi and doing data science into on-chain markets was MEV.
Joe: And MEV is essentially exploiting small inefficiencies in the pricing of orders in a block, and it's a highly competitive space. You might compare it to high-frequency trading in traditional finance. And this led me to this efficient market hypothesis through the work of someone called Eric Budish,
Joe: He's done some interesting papers comparing HFT and MEV. And I got very interested in this efficient market hypothesis, and that's really what's intellectually led my journey in DeFi and is still influences in a big way how I look at the market. So for anyone who isn't familiar, the caricatured efficient market hypothesis says that, everything out there in the market is already priced in.
Joe: Everything that is public information has already been factored into the pricing of yields or asset prices. So when you look at something and you say, "Oh, Tesla stock seems undervalued," probably you're just wrong. Probably you're just missing something because the market is actually efficiently pricing it, and the price of Tesla stock is fairly priced.
Joe: In which case you shouldn't buy Tesla stock because you have no edge. So that's one view. And a counter to this is to say if the market was perfectly efficient, then there would be no role for asset allocators and traders. The whole role of asset allocators is to try and identify where something has been mispriced and to go long or short or to take on a risk or to get exposure to some yield in some way.
Joe: And if the market is already completely efficient, then there would be no need for these people, but the actions of those people is what makes the market efficient. It's because traders are going long when they think something's underpriced that prices don't become horribly dislocated across the markets.
Joe: It's because people are arbitraging between different venues that assets tend to stay at the same price across those venues. So clearly there must be some level of inefficiency, in order to motivate these people to do the work that they do. And I got very interested in this question because my prior assumption was that the market is probably quite efficient in DeFi but the more that I started doing real empirical analysis, I was seeing inefficiency everywhere.
Joe: And when I saw these apparent inefficiencies, I assumed that probably I'm wrong about something, probably I'm missing something that the rest of the market is seeing and actually everything is fairly priced. But the more I worked on this, the more that I got involved in actively running DeFi strategies, I came to the conclusion that no, the market is genuinely quite inefficient.
Joe: Traditional markets I'm sure have some level of inefficiency, and I think DeFi markets have more inefficiency simply because it's a less mature market and it's somewhat more esoteric. In terms of where that inefficiency is, there is definitely more inefficiency on the long tail of things. So if you're looking at the largest deepest markets, it's very unlikely that you're going to find any edge there.
Joe: But if you don't look where the light is, if you look in some dark corners, then you might-- may find some huge inefficiencies that you can take advantage of. Now that's great. If you have a small account, I would urge anyone to, to do exactly this. Don't look where everyone else is. Look in the corners, look on obscure chains, obscure assets, and you'll find lots of things are mispriced.
Joe: But that is not very scalable. So then you, you-- when you start to get into large scale vault management you start to say, "Okay how can I exploit these market inefficiencies with large size?" And that's where the real challenge comes. There's really two sources that yield can be coming from.
Joe: Either you're transferring risk person A is just saying, "I will take on person B's risk for some price." Or you are actually identifying some kind of alpha. You're making a kind of arbitrage where you're receiving more yield whilst taking on the same risk as everyone else and that is inefficiency and in my opinion identifying market inefficiency is really the holy grail because that's where alpha can be found.
Nertila: I absolutely agree with you. In fact, this is why DeFi is getting a lot of traction. Due to the inefficiency traders have more edge to become more profitable.
Joe: Yeah, I would just add one final thing to that as well which is that edge can also come from access. So at Ember something that we're very keen on is the tokenization of real world assets and off chain yield strategies and I think that by bringing new products into this ecosystem by taking existing financial products, financial assets, and plugging them into this new pool of capital with all of these composable money Lego integrations that we can bring new sources of yield on chain and the people who have the first access to those new sources of yield will have some edge over those who don't.
Joe: So we're very keen on, on this at Ember bringing new yield primitives on chain.
Nertila: I want to say thank you for sharing your ideas, your insight with us today. Hopefully I will get you for another episode on our podcast in the future.
Joe: Thank you for having me on and I just wanna quickly say tradingstrategy.ai is a great website. Anyone who's looking into vaults and wants to compare performance, go and check it out. Tradingstrategy.ai indexes all of Ember's vaults so you can go in there, search Ember, and you can see for yourself the performance of our vaults and decide whether you want to deposit.
Nertila: Amazing. I love non-paid advertisement. Thank you so much.