I just fell down the most fascinating rabbit hole about Jane Street, and honestly, I need to share this because it completely reframes how I think about financial markets.

Here's the thing that blew my mind: a company with around 3,000 employees made $20.5 billion in net trading revenue in 2024. That's more than Citigroup's entire trading division ($19.8B) and Bank of America ($18.8B) combined. We're talking about a financial firm that almost nobody has heard of, doesn't have a CEO, and basically operates like what someone described as an "extremely profitable anarchist commune."

Let me break down who these people actually are.

It started in 1999 when three traders left Susquehanna International Group and a programmer from IBM decided to rent a tiny, windowless office in New York to trade ADRs (American Depositary Receipts). ADRs are foreign company stocks traded on the US market, and there are tiny price gaps between the US version and the original shares overseas due to time zones and information delays. These four guys—Tim Reynolds, Robert Granieri, Michael Jenkins, and Marc Gerstein—basically said "let's exploit that" and built algorithms to do it at scale.

The company was obsessively secretive from day one. No press, no conferences, no unnecessary exposure. Just quiet execution in a windowless room.

Then came the pivot that changed everything: ETFs.

In the early 2000s, ETFs were basically nobody's favorite market. They had low liquidity, few participants, and big institutions found them annoying to trade. But Jane Street saw exactly what everyone else missed—this was the perfect hunting ground. They became market makers for ETFs, meaning they post both buy and sell prices simultaneously and profit from the tiny spread between them. Sounds simple, right? Except it requires pricing thousands of assets at the millisecond level while managing insane inventory risk across global markets 24/7.

They did this with algorithms. And they did it better than anyone else.

What happened next was basically the textbook "picked the right wave" story. ETFs exploded from hundreds of billions to tens of trillions in assets. Jane Street became so essential to this market that you probably trade with them without knowing it exists. In 2024, they held 24% of the primary ETF market in the US, 41% of bond ETF trading volume, and 17% of European ETF secondary market share. Their average monthly stock trading volume hit $2 trillion. About 8% of all US options trading and over 10% of North American stock trading goes through Jane Street.

Every time you buy an ETF through your broker, there's a really high chance the counterparty is Jane Street.

Now here's where it gets weird: the culture.

Their headquarters has an actual Nazi Enigma machine sitting in the office—not as decoration, but as a statement. This company is obsessed with encryption, puzzles, and building its own isolated world. Their entire core trading system is built in OCaml, a functional programming language that basically nobody else in finance uses. As of 2023, their OCaml codebase exceeded 25 million lines—the Financial Times compared it to roughly half the Large Hadron Collider's codebase.

Why OCaml? Because in trading, one line of buggy code can cost hundreds of millions. OCaml's type system forces you to eliminate huge categories of errors at compile time. A side effect: engineers who work there become basically unemployable elsewhere because their OCaml skills don't transfer. It's an accidental moat.

There's no CEO. Seriously. About 30-40 senior employees make decisions collectively through management and risk committees. These people hold roughly $24 billion in equity and just... own the place. No hierarchy, no "Vice Presidents," no titles. The Financial Times called it an "extremely profitable anarchist commune" and honestly, that's not far off.

Employee compensation is tied to overall company profits, not individual trading performance. So nobody takes crazy risks for their own bonus because losses get shared by everyone.

That internship ad you saw floating around—$300,000 base salary, 4-month contract, no finance background required, no programming experience needed? That's real. But here's the catch: they only ask one question: Can you solve problems?

The interview process is notoriously brutal. Candidates get hit with probability problems, game theory puzzles, expected value calculations under pressure. They're testing your fundamental logic, not your resume. Only a tiny percentage of applicants even make it to the interview stage.

They don't use non-compete agreements either, which is almost unheard of in finance. Their reasoning: their competitive advantage isn't any single algorithm. It's the culture and capability density of the entire system, which can't be easily replicated.

Now, here's where things get really interesting: the people who came out of Jane Street.

Sam Bankman-Fried—yeah, that SBF—joined in 2014 with that $300,000 starting salary. He was exceptional at those probability problems. During the 2016 presidential election, Jane Street commissioned him to build a prediction system to know election results before CNN, then trade faster than everyone else. His system worked. They predicted Trump's Florida numbers minutes before CNN, realized the market was about to crash, and shorted the S&P 500 with positions worth several billion dollars.

When SBF went to sleep, Jane Street had a paper profit of $300 million. Largest single profitable trade in company history.

Three hours later, he came back to find the market had reversed. Trump was seen as pro-business, so the market rallied instead of crashing. Jane Street's short positions got squeezed. That $300 million profit turned into a $300 million loss overnight. A $600 million swing.

Here's the wild part: they didn't punish him. They basically said "your prediction model was accurate—the error was in market reaction judgment, which isn't purely mathematical." They even praised him internally. His salary went from $300K to $600K to $1 million bonus in year three. If he'd stayed, estimates suggest he'd be making $75 million annually by now.

But he left to start Alameda Research and FTX.

After FTX collapsed, it turned out Jane Street's alumni network basically dominated the entire drama: SBF (trader, 2014-2017), Caroline Ellison (Alameda CEO, former Jane Street trader), Gabe Bankman-Fried (SBF's brother, briefly at Jane Street), and Lily Zhang and Duncan Rheingans-Yoo (founded Modulo Capital with $400M from Alameda). The density of this circle is impossible to ignore.

But here's where things get darker.

In February 2024, two Jane Street traders—Douglas Schadewald and Daniel Spottiswood—quit and jumped to Millennium Management. Jane Street sued them and Millennium in April, claiming they stole a "highly valuable" proprietary trading strategy. What was this strategy? A short-term index options strategy targeting the Indian options market that generated over $1 billion in profits for Jane Street in 2023 alone.

After these traders left, Jane Street's Indian market profits dropped 50% in March 2024, while Millennium's Indian operations suddenly expanded. The case settled in December 2024 with confidentiality agreements, but by then the damage was done: Jane Street's disclosure of this "$1 billion Indian options strategy" caught the attention of India's Securities and Exchange Commission (SEBI).

SEBI launched an investigation, and what they found was stunning.

According to SEBI's 105-page ruling in July 2025, here's what Jane Street's algorithm allegedly did: On Bank Nifty options expiration dates, the algorithm would buy massive quantities of Bank Nifty stocks and futures right after market open (9:15-11:46 AM), sometimes accounting for over 20% of total market trading volume. At the same time, they'd establish large short positions in the options market—selling calls and buying puts.

Then in the afternoon (11:49 AM to close), they'd reverse: selling off all those stocks and futures they'd bought, artificially pushing the index down. The closing price on expiration date would shift lower, and their short option positions would generate huge profits.

On one specific day SEBI examined, Jane Street lost $7.5 million on spot and futures trading but made $89 million on options. Net profit: $81.5 million.

From January 2023 to March 2025, SEBI calculated Jane Street's total profits across all trading segments at 365 billion rupees (roughly $4 billion). But here's the kicker: they made 432.8 billion rupees in options trading while taking a 72.08 billion rupee loss in futures. The pattern was too clean to be coincidence.

SEBI's statement was brutal: "This outrageous behavior, which blatantly ignores clear warnings from NSE in February 2025, fully demonstrates that Jane Street is not a bona fide market participant and is not worthy of trust."

Context: 93% of retail options traders in India lose money annually, with losses exceeding 1 trillion rupees. Meanwhile, professional firms like Jane Street were printing money during the same period.

On July 4, 2025, SEBI suspended Jane Street from all trading in India and froze their accounts. On July 14, Jane Street deposited about $560 million into escrow and applied to resume trading. SEBI allowed it on July 21, but with ongoing investigation conditions.

Jane Street denied everything in an internal memo, calling SEBI's charges "highly inflammatory" and arguing their activities were legitimate arbitrage trading. They filed an appeal. As of now, the case is still pending.

But there's another lawsuit that's even more explosive.

In May 2022, Terra and Luna collapsed. UST went from $1 to worthless, Luna fell from $116 to near zero, and $40 billion evaporated. But four years later, there's a new theory about what actually happened.

On February 23, 2026, Todd Snyder (Terraform Labs liquidator) filed a lawsuit against Jane Street in Manhattan federal court. The centerpiece: a private chat group called "Bryce's Secret."

Bryce Pratt was a Jane Street employee who used to work at Terraform as an intern. He maintained connections with two people still at Terraform: a software engineer and a business development manager. The group chat was created in February 2022 and basically became an information pipeline from Terraform's internal operations to Jane Street.

On May 7, 2022, at 5:44 PM, Terraform quietly withdrew $150 million of UST from Curve's liquidity pool. This wasn't announced. Nobody outside knew.

Ten minutes later, a wallet associated with Jane Street withdrew $85 million of UST from the same pool.

Together, they pulled $235 million from that liquidity pool, directly breaking UST's liquidity support. The peg started failing. Panic spread.

According to Bloomberg's reporting on the lawsuit, Jane Street's actions let them "cover up hundreds of millions of dollars in potential exposure hours before the Terraform ecosystem collapsed."

Two days later, on May 9th, UST was already at $0.80. The collapse was irreversible. Bryce Pratt then messaged Do Kwon via group chat, suggesting Jane Street "could consider buying Luna at a significant discount."

It's like moving your valuables out during a fire, then asking the homeowner if they want to sell the rest at a loss.

Defendants include Pratt, Jane Street co-founder Robert Granieri (the only original founder still there), and employee Michael Huang. Jane Street's response was short: "A desperate lawsuit, transparent extortion." They added that Terra and Luna investors' losses stemmed from "billions in fraud" by Do Kwon and Terraform management.

That's technically true. Do Kwon did plead guilty and got 15 years. Terraform paid a $4.47 billion fine. But "Do Kwon is guilty" and "the others are innocent" don't automatically confirm each other. The building had structural defects. Whether someone removed the valuables beforehand during its collapse is a separate legal question.

So what exactly is Jane Street capital?

It's hard to summarize in one sentence. It's the most profitable trading firm on Wall Street—$20.5 billion in 2024 speaks for itself. It's a company with an impenetrable culture, a unique tech stack, and an acceptance rate that makes Ivy League schools look accessible. It's also apparently a deep player in legal gray areas, based on SEBI's 105-page ruling, Terraform's lawsuit, and Millennium's confidential settlement.

It could be all of these simultaneously.

Financial markets always have information asymmetry. What makes Jane Street unique is that they've weaponized this asymmetry to a systemic level. Michael Lewis wrote: "On Jane Street, a good trader isn't truly good unless they can clearly explain why they are good."

The core questions they seem obsessed with: What's the true market price right now? Where are the pricing discrepancies? How do we find and trade faster than everyone else?

The math questions in their interviews are puzzles. The Terra collapse is a puzzle. The disappearance of Bitcoin's "10 o'clock crash" after they got sued is also a puzzle.

Jane Street calls itself "a collection of puzzle solvers."

But when the market's attention finally turned to Jane Street itself, it too became a mystery. And that's probably exactly how they prefer it.