TechPrufer
Get started
Jane Street
Jane Street logoJane Street · Feature trail

Intern Projects Showcase

4 posts · Sep 2026 – Oct 2026

What this trail covers

This post showcases a variety of projects completed by Jane Street interns across different engineering and research domains, highlighting their contributions to areas like ML research, software engineering, Linux engineering, and trading desk operations. This edition specifically details the development of a centralized configuration management orchestrator for Windows, replacing a distributed PowerShell system with a more robust OCaml and F# based solution, including an audit mode for graceful migration and discrepancy detection.

All Jane Street trails
  1. Building and testing a config management orchestrator for WindowsThis post details the development of a new centralized configuration management orchestrator for Windows, replacing a distributed PowerShell system. The new system uses OCaml for parsing meta-config files and F# for stateless agents communicating via RPC to a central Orchestrator service. A key contribution is the implementation of an 'audit mode' in the client to detect drift between the old and new systems, which helped identify and fix bugs in the legacy system. It also describes a phased migration strategy.
  2. A FUSE server for our in-house storage backendThis post details the development of an OCaml-native FUSE library and a high-performance read-only FUSE server for the in-house Depot object store. It covers the implementation of the FUSE protocol, handling of Depot's object store semantics within POSIX filesystem constraints (including name length mapping and decoupling inodes from bundles), and performance optimizations such as parallelizing reads and reducing data copying by passing buffers directly, achieving 1.3 GiB/s read speeds.
  3. What the interns have wrought, special jumbo 2026 editionThis post details several intern projects, including an event-level generative model of market data using autoregressive diffusion, methods for combating memorization in LLMs, an activation checkpointing scheme outperforming PyTorch's for memory-constrained training, a performance testing framework with an indexing strategy for message framework tip recovery, an OCaml-native FUSE library for a high-performance object store, and a resilient kernel log reporter. It also mentions projects on sequence models for order book predictions, speeding up LLM prefill, bf16 vs. fp32 hidden-state degradation, interning-based compression of debug graphs, ILP/SAT solvers for scheduler assignment, a better coordination protocol for Gurobi clusters, and performance improvements in a feature engineering library.
  4. What the TDOE interns have wrought, 2026 editionThis post details three intern projects: 1. A system to compare internal voice trade records with public reports and flag discrepancies, handling issues like split prints. 2. Analysis of exchange specifications for 'Primary Peg' order types on MEMX and IEX, implementing the attribute in OCaml and FIX. 3. A tool to consolidate ETF closure calculations by considering market holidays across global constituents, with a UI to display fund placement, trading, and settlement availability.