Jane Street
Programming Languages

Jane Street is famously known for its deep commitment to OCaml — a functional, statically typed language that powers the majority of their trading systems, research tools, and internal infrastructure.

OCaml at Jane Street

• Type Safety: Strong static typing catches bugs at compile time, critical for financial code.
• Performance: Native code compilation delivers near-C speeds.
• Expressiveness: Algebraic data types and pattern matching enable concise, correct code.
• Open Source Contribution: Jane Street is OCaml's largest open-source contributor, maintaining Core, Async, and hundreds of libraries.

Why Not Java or C++?

Jane Street evaluated many languages and found OCaml's combination of safety, performance, and expressiveness uniquely suited to their needs — particularly for modeling complex financial instruments and market microstructure.

Official resource: https://opensource.janestreet.com/

Jane Street's OCaml Ecosystem

• Core: Their standard library replacement with better APIs.
• Async: Cooperative concurrency for high-throughput I/O.
• Incremental: Self-adjusting computation for reactive systems.
• Jenga: Their build system (now replaced by Dune).
• Dune: The widely-used OCaml build tool Jane Street created.

Practical Benefits

• Refactoring: The compiler catches breaking changes across millions of lines of code.
• Testing: Properties are expressed in types, reducing test surface area.
• Code Review: Concise, readable code speeds review cycles.

Learning OCaml

Jane Street runs an annual summer internship with OCaml training. Their book "Real World OCaml" (free online) is the definitive resource.

Real World OCaml: https://dev.realworldocaml.org/

While OCaml powers production systems, Python is heavily used for research and data science at Jane Street.

Python Use Cases

• Research: Exploratory data analysis, backtesting prototypes.
• Machine Learning: Model training and evaluation pipelines.
• Data Pipelines: Ingesting and cleaning market data.
• Jupyter: Interactive research notebooks.

OCaml + Python Integration

Jane Street builds Python bindings for their OCaml libraries, allowing researchers to prototype in Python and productionize in OCaml.

Key Python Stack

• NumPy, Pandas: Data manipulation.
• scikit-learn, PyTorch: Machine learning.
• Matplotlib, Plotly: Visualization.

Resource: https://www.python.org/doc/

For ultra-low-latency market making and co-location systems, Jane Street uses C++ where nanosecond-level performance is required.

C++ at Jane Street

• Market Data Feeds: Processing millions of market data events per second.
• Order Execution: Ultra-low-latency order routing.
• FPGA Interfacing: Hardware acceleration integration.

Modern C++ Practices

• C++17/20 features for safer code.
• Lock-free data structures for concurrency.
• Cache-aware data layout for performance.

The OCaml-C++ Bridge

Jane Street wraps C++ performance-critical components with OCaml FFI (Foreign Function Interface), combining correctness guarantees with raw speed where needed.

Resource: https://isocpp.org/

R is used for statistical modeling and econometric research at quant firms including Jane Street.

R Use Cases

• Econometric modeling.
• Statistical hypothesis testing.
• Time series analysis.
• Portfolio optimization research.

R vs Python in Finance

• R: Better for statistical modeling and academic research.
• Python: Better for production ML pipelines and integration.

Modern R Stack

• tidyverse: Data manipulation and visualization.
• quantmod: Financial data and charting.
• PerformanceAnalytics: Portfolio analysis.

Resource: https://www.r-project.org/

Top quant firms use a polyglot approach — each language chosen for where it excels:

• OCaml: Production systems, correctness-critical code.
• Python: Research, ML, rapid prototyping.
• C++: Ultra-low-latency execution.
• R: Statistical research and modeling.

Learning Path for Aspiring Quant Developers

• Start with Python for data science and finance.
• Learn C++ for systems programming.
• Study functional programming via OCaml or Haskell.
• Master statistics and linear algebra.

Key Resources

• Real World OCaml: https://dev.realworldocaml.org/
• Jane Street Open Source: https://opensource.janestreet.com/
• Quantitative Finance resources: https://www.quantlib.org/