Inside the $32B FTX Collapse: Ex-Employee Adi Baradwaj Reveals All

Introduction

The FTX collapse sent shockwaves across the cryptocurrency industry, leaving investors and employees devastated. In this revealing interview, Adi Baradwaj, a former engineer at Alameda Research, shares his firsthand experience working alongside Sam Bankman-Fried and what led to the $32 billion collapse. Adi provides unprecedented insights into the company culture, warning signs, and the moment he discovered his life savings were gone.

Founder Success Story QnA

Why now? Why did you decide to tell your story about FTX?

I thought about going more public with my story shortly after the collapse, but it didn’t seem like the right time with everything going on. All of us were really caught by surprise – we had no idea about all the stuff that had been going on behind the scenes. We were told about what had been going on by Caroline in basically a confession. Processing that took some time. It took me some time to just chill for a few months and collect myself and think about what had happened. It’s been long enough now at this point that I feel a bit more comfortable speaking out. Also, the trial is coming up, and I feel like it’s my responsibility to in some sense make some things known in terms of my experience at this company.

How did you end up working at Alameda in the first place?

I graduated from Berkeley in 2019 and went to work at Google for a few years. I wanted to leave to work on something a bit newer and more exciting, so I applied to a bunch of places, a lot of trading firms. One place was this small trading firm that no one had heard of – this was back in mid-2021. I’d only heard of Alameda because of a friend of mine from Berkeley who was mutual friends with Nishad Singh, who ended up being head of engineering at FTX. That helped make the intro to the company. I did the interview process, and all the people I spoke to seemed really intelligent, really motivated, really nice people. It seemed like a company with a mission.

What was your first month like at Alameda?

The first month was just like onboarding with any company – getting up to speed on the tech stack, getting used to the actual object level work, how to get stuff done at the company. My first month was in the office at Berkeley, next to campus. That was where the company started in Berkeley in like 2017.

Who was Sam Bankman-Fried at that point? Was he already famous?

There had been some coverage of FTX and Alameda in the news, but nothing like what it came to be a few months after that. It was like the typical Tech Crunch launch people do when they raise money. People knew about the existence of FTX, and it was a big name within crypto, but if you’re not in crypto and you’re not in trading, you probably hadn’t heard of it at all.

What was this weird relationship between Alameda and FTX?

They were separate companies, probably different legal entities, but socially the companies were very mixed. I got to meet several employees of FTX socially. We had joint hangouts, joint social events. Employees sometimes lived in the same housing spaces in the Bahamas and also in Hong Kong. There were kind of fuzzy boundaries, which I think is a larger symptom of the fact that Sam and Caroline and the other folks did not really pay close enough attention to separating things and making things proper in the way that it should have been for two firms of that size handling that kind of trading volume and that number of users.

What were the mission differences? What was the actual differences between Alameda and FTX?

The actual work of the companies is completely different. Alameda was a trading firm – we did all kinds of trading, market making, standard statistical arbitrage across exchanges and on individual assets across time. FTX was an exchange – a place that people go to actually perform their trading, buy and sell various crypto assets. FTX runs a matching engine that matches these trades together and actually executes those trades and settles those trades. They’re two very different businesses but in the same space of crypto trading.

Were you able to tell that there was something going wrong or was there something off about Sam?

Sam’s a weird guy. He has a unique personality. If you watch any of his interviews, you get a sense for the way he talks, the way he thinks. He’s unique in many ways – he was the founder and majority owner of these two multi-billion dollar companies that he managed to build in the span of a few years. For me and for a lot of people at the company, we saw that and thought, “Oh, this guy’s a genius, he’s cracked the code, he’s figured something out that nobody else has been able to figure out in the startup community.” What added to that was the fact that he had this effective altruism mission, which was a very compelling idea.

What is effective altruism?

The idea is that there are a lot of ways you can do good, and we should be thinking about which of those ways are most effective as measured by things like number of lives saved relative to each marginal dollar donated, or how many quality-adjusted life years you’re adding for each action you’re taking. Concretely, folks like Sam and Caroline were really into the idea of making a lot of money and then donating that money to effective causes. They talked about it a lot both internally as well as externally, and it was a big part of the DNA of both companies.

Do you think that helped when raising money and gaining trust?

That’s what a lot of investors thought. There’s a funny article that Sequoia Capital posted right after they made their investment in FTX. Reading through all the praise they’re heaping on Sam is just something else. The thing about him playing League of Legends while taking the investor call is a classic example of seeing something a certain way beforehand and then realizing you were completely wrong. All of us had that realization, and it was a hard one because we all paid a lot for it – financial loss, wasting years of our lives dedicating time and energy toward this company and person and cause.

When did you start finding out things were broken?

There were little things that were strange and kind of silly. The company spent a lot of money on things they didn’t need to. For example, our housing in the Bahamas was incredibly opulent – we lived in five-star accommodation. A bunch of employees had space in Albany, which is a really fancy resort in the Bahamas where a lot of celebrities live. We had a recording studio next door where Drake and Oprah live. That’s the kind of thing where you think, did we really need to shell out that much money? There were many more dumb decisions that the companies made that didn’t need to have been made.

What were some of the things you got to do while you were on the team?

The travel was nice – being able to travel between offices, living in this beautiful island country. The environment was amazing, and I worked with incredible people. I learned a ton at the company, a lot that I’m even using now at my current company. There were funny incidents – one I’m planning to tweet about happened the week I joined, when a trader at Alameda fat-fingered a trade. They typed in the wrong number, sold a ton of Bitcoin at a really bad price, and there was a flash crash from maybe 50 or 60k all the way down to 8k on Binance. The company lost tens of millions of dollars.

What were some of the main key events that made FTX crash?

There’s really one key event. People will say CZ from Binance selling FTT caused the crash, but that didn’t cause it. What caused the crash is the fact that Sam took 8 billion of customer funds and used that to bankroll a bunch of risky venture investments that were outside of Alameda’s and FTX’s core business without telling anybody at the company except for his core group.

Why would he do that? What’s the motivation?

You have to ask him – unfortunately, he’s behind bars now. We can speculate. In a lot of his interviews, Sam talks about this idea of linear wealth preference. One of Sam’s core beliefs was that you almost always take the double-or-nothing bet. He believed that the way you act in terms of risk with small amounts of money is the same as the way you should act with large amounts of money. The fatal flaw in that logic is that sometimes you’re going to lose a lot, and when you do something like that and go to jail, you don’t get to play the game anymore.

What was it like for you when Caroline confessed about the company going down?

It was wild. All of us were just sitting on the floor in a circle. Caroline was on a bean bag, crying a little bit. It was a very tense atmosphere. As soon as we realized the scale of what she was talking about, the tone immediately changed. At first, we were trying to understand, but once it became clear that she had done things along with Sam that were probably criminal, it was not a friendly conversation anymore. We grilled her with questions – about our equity, our deposits on the platform, whether we’d get paid, whether we should leave the country, who would pay for flights.

What was going on in your mind as you left that room?

Right after that meeting, all of us except Caroline went out for drinks. We were trying to process what was happening and needed to talk with each other. We were wondering what to do next – go back home, talk to lawyers about our exposure. We weren’t really worried about legal exposure, but we wanted to help with the investigation process as much as possible.

What happened after that? When did you fly home?

We all handed in our resignations the following day, packed our stuff from the office, and separately booked tickets to go back home since we were all from different places. I flew from Hong Kong to India, spent some time with family, then came back to San Francisco. I still have belongings in the Bahamas that I don’t know what happened to.

Did you tell family and friends? Were there NDAs?

There was no NDA with the company, and the company doesn’t exist anymore. I told family and friends about it. After coming home, I took a few months to chill and rewind and process what happened. I followed the case and all that stuff in the media. Former employees have a couple of group chats we stay in touch with.

Do you still believe in decentralization after what happened?

I do believe that decentralization is the future. There’s so much genuine value in the space – a lot has been realized already but been obscured by some of the stuff that’s happened recently like FTX. There’s a lot that hasn’t even been realized yet. I’m very optimistic about the space, about crypto, about decentralization, and that’s why I’m building in this space.

What are you building now?

It’s called Astra. We’re building a universal connectivity layer for crypto markets. Currently, if you want to buy or sell traditional assets, it’s straightforward with established venues. But in crypto, you have hundreds of trading venues – centralized exchanges, decentralized exchanges, protocols for trading and lending, OTC desks, custodians. It’s a huge zoo. We’re building infrastructure to integrate with all these venues – robust, low-latency connectivity. For users, it’s just an API where you can access Binance, Coinbase, Uniswap, dYdX, any venue you want to trade on, using the same data format.

Why has no one built this yet?

It does exist in fragmented ways, but there’s no satisfying solution that builds extensive connectivity – integrating with not just one or two venues but 100 trading venues, which is the reality of crypto. People trade on all these different venues, and we don’t know which L1 will end up winning or even if it’s going to be an L1. The infrastructure to integrate with these venues isn’t built out in a satisfying way for businesses and individuals.

How are you applying your FTX experience to building Astra?

We’re solving the problem by saying we’re not going to handle custody or counterparty risk. We give a direct market access to trading venues. Our customers come with their own API keys from exchanges they’ve already onboarded with. It’s like Plaid for crypto – bring your own API keys, bring your own accounts, bring your own KYC. We’re selling software to businesses, and those businesses have full control of all their data and keys.

What’s it like being a founder now after working for SBF?

It’s extremely fun and exciting. It’s not one hat, it’s like a hundred different hats I’m having to wear every day. Swapping between them and working on all these different roles within the company is amazing. The origin of this company is really in a lot of the problems I saw at Alameda in terms of our technical infrastructure. We’re trying to solve those problems directly related to stuff I’ve done in the past. I’m thrilled to be building this – it’s exciting every day when I wake up and get to work on it.

Where’s the product at right now?

We’re building the V1. We have integrations with about five centralized exchanges, three DEXes, and we’re working on one custodian. We’re testing to make sure it’s super solid and robust. The plan is to have a Beta release out in two months. We have some customers we’re talking to, and if anyone sees this and would like to get in touch, we’d absolutely love to talk to them.

How should people reach out to you?

Our website is Astra API dot dev. We’re also on Twitter at Astra API. You can find me there – my name is Adi. We’d love to hear from potential customers or anyone interested in what we’re building.

Adi Baradwaj Business Stats

Adi Baradwaj is currently building his startup Astra after the FTX collapse. Here are some key stats about his current venture:

  • Funding: Recently raised a $2 million seed round
  • Team size: 5 people (combination of past colleagues and friends)
  • Current focus: Building V1 product with exchange integrations
  • Target market: B2B SaaS for crypto businesses needing trading connectivity
  • Launch timeline: Beta release planned within 2 months
MetricValue
Company NameAstra
Founded2023
Seed Funding$2,000,000
Team Size5 employees
Business ModelB2B SaaS

Adi Baradwaj Method

Adi’s approach to building Astra is directly influenced by his experiences at Alameda Research and witnessing the FTX collapse. Here’s how he’s applying those lessons:

  • No custody of user funds – avoiding the risk that brought down FTX
  • Bring-your-own-keys model – ensuring users maintain control
  • Focus on infrastructure rather than speculation
  • Building traditional B2B SaaS without token offerings
  • Transparent communication with users and stakeholders

Adi Baradwaj Tools

Adi is leveraging his technical background to build Astra’s connectivity layer without making the same mistakes as his former employer. The company focuses on providing essential infrastructure for crypto markets.

  • API connectivity layer for multiple exchanges
  • Low-latency trading infrastructure
  • Unified data format across different venues
  • Security-first architecture with user-controlled keys
  • Traditional B2B SaaS tools for customer management

Key Notes

The FTX collapse offers important lessons for anyone in the crypto industry. Here are the key takeaways from Adi’s insider perspective:

  • Warning signs were present but overlooked due to SBF’s perceived genius status
  • The effective altruism mission was used to build trust while masking fraudulent activities
  • Proper separation between companies is crucial, especially when handling customer funds
  • Extravagant spending should be questioned even in successful companies
  • Transparency and proper risk management are non-negotiable in financial services

Get Started in Just 5 Steps

Based on Adi’s experience building in the crypto space post-FTX collapse, here’s how entrepreneurs can build trustworthy crypto companies:

  • Prioritize proper business structure and separation of concerns from day one
  • Implement robust risk management systems before scaling operations
  • Build transparent, auditable systems that regulators can inspect
  • Focus on solving real problems rather than hype and speculation
  • Maintain open communication with users and stakeholders about business practices

Conclusion

The FTX collapse serves as a cautionary tale for the cryptocurrency industry. Through Adi Baradwaj’s insider account, we’ve gained unprecedented insight into the culture, decisions, and personalities that led to this $32 billion disaster. Despite the trauma of losing his life savings, Adi remains optimistic about crypto’s future and is building Astra with the lessons learned from this experience. His story reminds us that while innovation is crucial, proper governance, transparency, and ethical leadership are non-negotiable foundations for any sustainable business, especially in the financial sector.