Formal Methods Research and Development

About the position

We're opening our first internships in the Formal Methods group of our OCaml Language team, which develops language features and tooling for correctness reasoning and bug finding in OCaml programs. With the emergence of agentic programming, we think formal methods will play a major role in helping validate agent-generated code and providing feedback for agents. Our vision is to make formal methods as useful a tool for building software as type systems are for us today, both by leveraging existing tools and technology, and pursuing novel approaches. 

During the internship, we want to give you a real sense of what it's like to work at Jane Street full time while also providing a truly unparalleled educational experience. You'll be paired with full-time employees who act as mentors, collaborating with you on a real-world project that combines your research background with one of our active projects, such as adding support for property-based testing of formal specifications or developing a refinement type system for functional correctness verification. You’ll have the opportunity to evaluate the tools you develop on a large, actively developed production codebase. You'll also be able to attend guest speakers and social events, and hopefully get a real sense of what it would be like to work here full-time.

There is plenty of work on combining agents and formal methods. What’s different at Jane Street is that we aren't stuck designing verification tools for legacy languages: we can modify and improve OxCaml, our branch of OCaml, and its development environment to make them well-suited for proof-driven software development. Moreover, we have a community of excellent programmers who are eager to adopt these tools and apply them to real software.

Learn more about Jane Street's internship program here.

About you

We're looking for PhD students, postdocs, and masters students with outstanding research experience in program verification, programming language semantics, property-based testing, static analysis, model checking, and related areas. You should be

• A strong programmer who enjoys collaborating in small teams
• Excited to turn research ideas into practical tools
• Intellectually curious, humble and unafraid to ask questions 
• Fluent in English

We don't expect you to have a background in finance, OCaml, or functional programming.