Some quick, important, contextual pieces:
A little over a year ago, marginfi was born. The first half of 2023 was a time when no one believed in Solana. No one.
Investors didn’t fund new companies here, economic activity was small here, and public sentiment was in the trash. It’s crazy to think about this now, but Solana was in a completely different spot a short year ago. Things can change quickly.
However, the team at mrgn was convicted in the optimizations this “contrarian” tech stack could provide. We were bullish on the teams that surrounded us. We saw a pack of differentiated users that wanted to do more with their holdings on-chain. We saw a ton of dormant, bottom-of-funnel liquidity (vanilla staked SOL) that we knew would work its way into LSTs (part of the reason we built $LST) and we doubled down.
We launched marginfi and immediately started connecting with power users. marginfi finally gave them a powerful, new lending market focused on risk management with a completely new interface. We prioritized “blue chip” tokens and long-tail tokens, with one of our first listings being BONK. People finally had a way to go levered long (and levered short) with low-liquidity, long-tail assets on marginfi, while still isolating risk from main pool lenders. This is a big part of Solana’s usage today, and we’re soon releasing a product to make this even easier.
This growth wasn’t enough. A lot of retail traders had become very pessimistic on Solana because they thought they couldn’t make outsized returns here. One of the main criticisms with Solana is that “nobody made money there”. We were set on changing that narrative and showing that user engagement mattered. What we did next shaped a new meta for crypto.
In July 2023, we launched the first defi points program in crypto. This enabled us to be more public with the ways we indexed user engagement on Solana. Points accumulated based on linear capital engagement, which also transformed the ways a lot of teams thought about sybil prevention and what really matters to a protocol.
At this point, we were already the fastest growing defi protocol in crypto. However, we still saw a lot of locked liquidity in vanilla SOL stake. To unlock this, we shipped $LST — the highest naturally yielding Solana liquid staking token. $LST delegated to mrgn validators which run 0% commission, also taking no LST fees, and we incorporated a plethora of voting modifications to our validators, which significantly outperform the field in earning vote credits (and delivering yield to holders). mrgn validators also run the Jito-Solana client, earning MEV captured within that environment. We gave SOL stakers seeking the highest return an effective way to make that stake liquid and use it as collateral in marginfi.
The last big jump we made was a completely new, mobile first lending experience. For the first time, users could download marginfi to their phone (from the web), access any asset through a single tab, and sign in without a wallet using email, Google, Twitter, or Apple. You can use marginfi like you use a normal application, except for one last thing: being able to get marginfi from the App Store or Google Play. Don’t worry, we’ll fix that too :)
marginfi finished 2023 as the fastest growing defi protocol in the industry that year.
Current Focus
2024 began, and our focus shifted to a few core things:
mrgnlend robustness
new products
application growth
mrgnlend robustness points to a lot of workstreams around risk management, our developer and user ecosystem, functionality enabling the application layer, and protocol & application security. marginfi has a lot of liquidity in it, a lot of borrows, a lot of users, a lot of different assets, and a lot of stakeholders (contributors, liquidators, ecosystem developers, investors, etc).
We don’t take this lightly at mrgn — we’ve been deeply focused on additional audits/security testing, we’ve been automating our risk pipelines and diving into deeper testing simulations, we’ve been reinforcing our liquidator network with strong independent searchers, teams, and our own liquidator infrastructure, we’ve been working closely with mrgn ecosystem teams on new solutions enabled by marginfi’s architecture, tooling (like flashloans), userbase, and liquidity, we’ve been working on critical platform infrastructure that can get through severe Solana congestion and inefficiencies, we’ve been pushing to make marginfi more accessible to non-crypto users with our mobile, wallet-less application, with features like collateral repay and flashloan swapping, and we’ve been working with Anza, Jump, and Jito on core Solana improvements to prevent the next major technical hurdles to Solana usability.
This is a massive push on our end, and a lot of this work flies under the radar. Just because some of it is less visible doesn’t underscore its importance.
We’ve also been working on some new products, and I have updates. We started 2024 building YBX, a decentralized stable-asset backed by LSTs. We believed a lending protocol, CDP-type stablecoin, and stableswap are essentially the same product. We still believe this, but defi isn’t a place where you can launch something as precise as a stablecoin or lending-integrated stableswap without resilient supporting infrastructure in place, and a robust environment to grow these things on the back of a complicated, risk-centric protocol like mrgnlend. There are many things that need to be improved with Solana, many things that need to be improved on marginfi, and many things that need to be improved with supporting infrastructure (like oracles) for us to feel more comfortable pushing these new products to production.
Solana is still incredibly young, defi is still incredibly young, and it’s ok to admit this. If new product growth comes at the expense of predictable significant risk, we aren’t doing our job protecting users. So we’re spending considerable time on Solana improvements, on marginfi improvements, and working with supporting teams and infrastructure to make the environment for YBX and mrgnswap more stable, first. YBX is code complete and double audited. We could launch it right now. I know users want it, and we want to give it to you — but hopefully you can better understand our perspective and the responsibility we place on ourselves for everything we build with this specific update.
We are about to release a different product, however. We’ve been quiet about this one, but I expect it to turn some heads. It’s completely new, and it’s something that will unlock a ton of liquidity for trading spot assets on Solana — especially all your favorite memecoins. It’s something users want, and marginfi will be the only place to do it. As always, Twitter, Substack, and Discord will hear about it first.
We also recently announced our focus on becoming Solana’s liquidity layer for any current or new defi team. We’ve picked up significant traction on this. There’s now 15 strong teams building differentiated, completely new solutions on marginfi. marginfi offers developers something no L1 does: immediate liquidity, a strong userbase, distribution, and a host of off-chain tooling — all of which is getting improved every day. You can’t copy/paste on marginfi — it’s not just another EVM compatible L2. That means all 15 teams are building intentionally on marginfi for the robustness we’ve engineered and the resources we’ve aggregated. We’re actively rolling out a grants program around this and we’re meeting with new teams everyday that want immediate benefits. Let’s get in touch if this is you.
The other big workstream we’ve been focused on in 2024 is application growth. Crypto applications suck. Imagine asking your mother, father or friend to use a lending protocol on Ethereum. They’d have to find a way to transfer money from their bank to buy Ethereum on an exchange, download a wallet, learn about self-custody, learn how to swap into the asset they want to lend, find where to swap, pay multiple, insane transaction fees, and then eventually find a way to get that crypto back into their bank account, which involves all those steps in reverse. This is slow, this is multi-step (and multi-product), this probably isn’t mobile, and they need to manage not to get hacked in between steps, which isn’t a easy task for a new crypto user.
Now take marginfi. You go to marginfi.com, open the app (download it from the web if you want), sign in with email or a social account, buy the Solana you need, perform a guided swap into whatever assets you want to earn incredible yield on, and boom - deposit. No insane gas fees, no third-party wallet, no exchange needed to initially buy your SOL… nothing. Just use the app, share it with friends, and get rewards for it. This is what crypto should be, so we’re building it. Borrowing is the same way: hold the assets you want while getting liquidity you need in the meantime — quick, cheap, fast, and all managed in a single dApp.
We’re about to launch a new onboarding flow that embeds these principles. I invite you to try it next week and to share marginfi with friends who can try it. If you’re a consistent marginfi user, you might need to reset your cache to try it (so you don’t get automatically signed in). marginfi is not just for the crypto natives, it’s for everyone. We’re building marginfi’s application for a global audience. A native mobile app that you can find in the App Store or Google Play is next.
The other big application push is our new trading app, which will be launching soon. This is the product I teased earlier in the article — there’s going to be some incredible things you can do with it that you can’t do anywhere else. We’re going to let this product do a lot of the talking for us.
Another new product, mrgnloop, will be launching next week. This is a product made easy by our flashloans and collateral repay. mrgnloop is a no-brainer for anyone looking for managed, straightforward LST rate arbitrage. It’s also a great way for marginfi to support greater LST adoption, unlocking the billions of vanilla stake that isn’t helping defi until it’s turned into LSTs — like $LST.
Collateral repay was another massive release making it simple to repay loans with a user’s existing marginfi portfolio. If you use a different lending venue, you probably have to go swap on a DEX whenever you don’t have enough existing inventory in your wallet to repay a loan. On marginfi, you never have to leave. You can use existing collateral to swap into the inventory you need and repay a loan, all using flashloans. There’s a ton of work that goes into seemingly small features like this, which is why marginfi is the only place on Solana much of this is available.
Our application growth has been extremely fast, and we’re only looking to accelerate. More on our product velocity later.
Wins, Losses, Criticisms, and Praises
marginfi has an extremely passionate, strong user base. mrgn mamas are demanding, they’re loud, and they want the best. We couldn’t ask for a better community. This community makes our wins big and our losses bigger. There’s been some notable wins and losses, and I want to highlight them here:
Losses:
In April, three things happened back to back (to back) that sparked commotion and controversy across the industry. 1) SolBlaze wrote a viral tweet about us containing material disinformation, which we disproved with on-chain evidence. 2) One of our cofounders left in a publicly damaging way. 3) Solana encountered unprecedented congestion, forcing us to loosen parameters on some oracles which led to 0.5% of our users getting incorrectly liquidated, which we refunded. A few thoughts here:
Crypto is extremely financialized and incentivized. People will attack others when they feel their invested interests are threatened. We see this with ETH and SOL, we see this with competing products, we see this everywhere. This plays into public narratives and shapes messaging, opinions, and sentiment. marginfi is easy to hate by default because it’s on Solana and doesn’t have a token yet. The three events in April opened up a stream of attacks on top of that. However, the marginfi protocol remained resilient through it all.
The main thing I want to stress from that period is that marginfi remains committed to our users, and we encourage all users to verify, not trust, in return. marginfi is open source, double audited, code verified, and completely transparent. Twitter narratives will always fall short of what’s happening on-chain. It’s incredible to see on-chain that marginfi growth immediately continued after the combined April incidents. We have an incredible upcoming pipeline, and half of 2024 is still left. We’re going for the fastest growing defi protocol in crypto two years in a row :)
Wins:
marginfi is now the most stress-tested lending protocol on Solana. On that day in April, during unprecedented Solana congestion, $300M was seamlessly withdrawn from marginfi as users stressed by Twitter narratives temporarily pulled money out. Yes, $300M in 24 hours. Your bank can’t handle that (you’d get your assets frozen and delayed). Most defi protocols can’t handle that (many have daily withdrawal limits, let alone their risk systems being prepared for these circumstances). marginfi facilitated it seamlessly. This means liquidators were liquidating, loan repayments were happening, oracles were getting data through, and transactions were landing even as users struggled making a basic swap through the Solana congestion. One other thing — this is while Solana price was plummeting. Most of marginfi’s collateral is in SOL and LSTs. Our risk systems handled SOL dropping from $200 to $126 effortlessly. I’m so proud for the resilience the team had during this period, and we have a badge of honor to show for it.
There’s a ton of other wins we had, but I want to focus the attention on this particular win. Always ask yourself hard questions about risk when using defi — remember that you are your own risk manager. When it was time to step up, marginfi stepped up.
Criticisms:
We get criticized every day and much of this is great. We can learn from our users and work to build them a better product. Some criticisms are hard to respond to, especially when they dive into legal territory. Nevertheless, I’ll summarize a few of the main criticisms that are harder to address transparently below:
mrgn Points. I talked about the reason we launched points earlier in this blogpost, but I’ll expand a bit on criticisms here. We launched points because we’re building a rewards campaign for marginfi users, adjacent to the world-class over-collateralized yield users earn on marginfi. After we launched Points, pretty much every team in the industry built in their own points system. Teams quickly made points solely about token rewards, and shifted user expectations. We’ve been clear since the start that mrgn Points are not directly tied to tokens. mrgn Points are one user engagement metric (of many) that we made public to show how we assess user capital contribution. Despite this, we recognize where user expectations are so we’ve been heads down thinking on ways to adapt our points system. Changes (and rewards) are coming.
Token. Users obviously want MRGN Foundation to launch a token. Let me be clear: marginfi is built to have a token. Nothing has changed. We strongly believe in tokens, we see tokens as a medium for governance, and we think a token is important for the current and future stakeholders of marginfi. However, launching a token prematurely or launching a token without protecting essential contributors to marginfi is not something we’re interested in. Having a token live means we need the critical resources, systems, and governance processes in place to effectively unify the marginfi token with the marginfi protocol. We also need to ensure marginfi is ready for that transition from a growth standpoint. marginfi needs deeply embedded distribution flywheels and effective contribution incentivization if it’s truly going to be a liquidity layer for global finance. We’re getting there, but there’s still a lot we need to do. More in the next criticism :)
Farming users. marginfi is just over a year old. It’s one of the newest protocols on Solana. I’ll often hear a user complaining that marginfi is farming its users. Sorry, but I respectfully disagree. Iconic products aren’t built in a year. Iconic products are more typically built across 10 years. That’s not an exaggeration — look at some of your favorite products and when they started to work and really scale vs when they were started. Crypto is the only place online where you have actual self-custody over your property. Users have ultimate optionality in crypto — more than anywhere else. This makes crypto great. At mrgn, we’re extremely focused on continually understanding why users choose marginfi and what we can do better to make that choice easier. We intentionally separate the marginfi protocol from incentives when we do this. Incentives aren’t the product. If a user is only interested in incentives, and the product makes no difference, you have no product. We’re here to build something completely new for the world. For the first time in human history, anyone with an internet connection can earn transparent yields with no middlemen, in whatever asset they want, without crippling fees, with deep liquidity, instantly. That’s a product with incredible potential that we want to get right.
Shipping speed. mrgn mamas want new products as fast as we can get it into their hands. We’ve done a pretty good job on velocity at mrgn. Find a team of 9 people that can ship everything I’ve detailed in this post safely in a year and I’ll give you a soda. We’ve raised $8M total, have had no serious security incidents, and have already been through incredible trails and stress testing in our short history on Solana. For those that want us to go even faster, I want to remind you that permissionless finance is mission critical. Changes need to be reviewed, risk needs to be managed, downstream effects need to be analyzed. User safety is the #1 priority at marginfi. Safety will slow our process down at times, but this is always to the benefit of the user. Defi is a dark forest we cannot control. We can only manage risk and provide our users with the tooling they want. This is fun, this is new, and this is why marginfi has the potential to change the world.
Praises:
We’ve had some incredible users, partners, and investors during our short history. We’ve onboarded quite a few people to Solana and to defi as a whole. We have some amazing teams building on marginfi that we can’t wait to dive deeper with. To all these people, thank you for your commitment and determination to grow this space. If you’re coming to defi, marginfi is the best place to earn transparent passive yields, to borrow what you need against your holdings, to mint inflation-resistant SOL (through $LST), and soon to have fun leverage trading any Solana token you want. Download marginfi from the web on your phone, sign in with email, and get started today — use a friends points reflink when you sign up to give them a points boost!
boolish on accessibility maxxing
Love the platform Mac. But.
If you had a board of management you would have been sacked by now.
You can’t keep dangling the carrot of a token and decentralised protocol indefinitely. You are pissing off everyone, including your team.
Decentralisation doesn’t have to be perfect out of the gate. Like a product, it can grow into the governance mechanism you want. But there are many great examples of starting small, delegating stake, voting on minor issues. You seem to want to overcomplicate it. Meanwhile your users are out of patience. These aren’t greedy farmers. They left ages ago when things blew up. These are rusted on supporters who have locked significant liquidity. Who want to be part of Margin for life! But you have given them no avenue to be part of Margins future. Great work but empty promises is your calling card.
I have 90% of my net worth in Margin, because i like the platform. But it only takes a good experience and an incentive for margin to lose that goodwill forever.
Please find the bandwidth to do a token, governance lite, and let people grow with the platform.
This really is your last chance. Sincerely.