Neural networks are a tool we use heavily in many trading strategies. But raw market data can’t be fed to a neural net as-is. It’s an unstructured input that needs normalization and feature extraction first.

The following is part of a series of posts about 2026 summer intern projects—for more, see “What the interns have wrought, special jumbo 2026 edition”

The following is part of a series of posts about 2026 summer intern projects—for more, see “What the interns have wrought, special jumbo 2026 edition”

Jane Street does a lot of network maintenance over the weekends. In each maintenance, a cabinet may lose connectivity for 15 to 25 minutes. Our internal Kafka infrastructure is supposed to be resilient to these partitions, and reconnect when connectivity resumes. For several months this year, it did...

The following is part of a series of posts about 2026 summer intern projects—for more, see “What the interns have wrought, special jumbo 2026 edition”

The following is part of a series of posts about 2026 summer intern projects—for more, see “What the interns have wrought, special jumbo 2026 edition”

It’s time again to review some of what Jane Street’s interns built over the summer, though this year we’re going to try to cover more projects and in more depth. In particular we’re including projects from teams and roles whose work has historically been hard to talk about publicly—like ML research ...

The following is part of a series of posts about 2026 summer intern projects—for more, see “What the interns have wrought, special jumbo 2026 edition”

The following is part of a series of posts about 2026 summer intern projects—for more, see “What the interns have wrought, special jumbo 2026 edition”

The following is part of a series of posts about 2026 summer intern projects – for more, see “What the interns have wrought, special jumbo 2026 edition”

The following is part of a series of posts about 2026 summer intern projects – for more, see “What the interns have wrought, special jumbo 2026 edition”

TL;DR: We study the scaling laws of data weighting across in-house and open-weight LMs, finding non-monotonic behavior across scales. We vary the weight assigned to sequences during training and measure how strongly the model’s loss reduction on a sequence depends on the sequence’s weight. Taken tog...

Jane Street is known for being an OCaml shop, but for years now Python has been our second major programming language, acting as the primary tool for data analysis and (especially importantly these days) machine learning. Most of our traders and researchers think and write in Python, even as the maj...

I’ve been telling people for the last 25 years that Jane Street as an organization was just not interested in formal methods.

We’ve always found strace useful but somewhat hard to work with. Its output is often inscrutable, it’s hard to follow subprocesses or threads, and if you want to filter syscalls you have to rerun the trace with a flag for each one. What you want in debugging is a tool for exploring, refining, etc., ...

Attention is a computational primitive at the core of modern language models, allowing internal representations to reference and influence each other. It’s how these models handle sequential data in the first place.

A lot of “capture-the-flag” style ML puzzles give you a black box neural net, and your job is to figure out what it does. When we were thinking of creating our own ML puzzle early last year, we wanted to do something a little different. We thought it’d be neat to give users a complete specification ...

At the end of last year, we decided to try something new: a challenge that would run alongside Advent of Code, where we asked the community to show us how they could design hardware to solve the same problems. We had no idea what level of participation to expect, but we received a huge number of sub...

For a long time I was skeptical of LLMs—whenever I reached for them I was disappointed by the results. Last year I tried Copilot and Cursor to tweak a game I’d built, and neither generated working changes. At a previous job I tried Gemini to outline product briefs and generate wireframes, but ended ...

I recently ported the Hardcaml_step_testbench library, one of the libraries that we use at Jane Street for Hardcaml simulations, from using monads to using algebraic effects, a new OCaml 5 feature. This blog post walks through what algebraic effects are, why you should consider using them in lieu of...

Testing is an essential part of building reliable software. It’s a form of documentation, a reminder of mistakes of the past, and a boost of confidence when you want to refactor. But mostly, testing is a way of showing that your code is correct and resilient. Because it’s so important, we’ve investe...

Update: We got over 200 submissions to this challenge, spanning a wide variety of HDL languages and hardware platforms! We featured our favorite submissions in the results blog post, check it out here

Yet again, we’re at the end of our internship season, and so it’s time to summarize what the interns were up to!

We recently ran across a strange higgs-bugson that manifested itself in a critical system that stores and distributes the firm’s trading activity data, called Gord. (A higgs-bugson is a bug that is reported in practice but difficult to reproduce, named for the Higgs boson, a particle which was theor...

At Jane Street, we’ve been actively making improvements to OCaml for a long time. Over the last few years, we’ve started to build some fairly ambitious extensions to the language. Our aim is to make OCaml a great language for performance engineering. This work has always been open source, and our ho...

Update: For the 2025 Advent of Code, we ran an Advent of FPGA Challenge where we invited the community to implement their own synthesizable solutions to this year’s puzzles! We got over 200 submissions spanning a wide variety of HDL languages and hardware platforms, check out our favorite solutions ...