ML Compiler Engineer - High-Throughput, Low-Latency ML

Jane Street; Greater London; On-site; GBP 90,000 - 140,000; Full time

Job summary

Jane Street seeks compiler engineers to join a new ML Compilers Team in London. You will help translate research ideas into high-throughput training and low-latency inference, shaping the compiler stack and APIs used across ML workloads.

The role involves collaborating with researchers and engineers to optimize graph captures, DSLs, and kernel execution on GPUs, while maintaining clean interfaces and robust tooling.

Qualifications

• Experience on ML compilers or related stacks
• Knowledge of ML frameworks and graph compilers
• Comfort with English and strong communication

Responsibilities

• Shape the development of Jane Street's ML compiler stack.
• Collaborate with ML researchers to translate research ideas into production systems.
• Identify and implement performance optimizations for compilation and inference.
• Contribute to frontend/backend components and tooling for the compiler pipeline.

Skills

ML compiler

ML frameworks

Graph compilers

Kernel DSLs

GPU programming

SSA/CFGs

CUDA

Tools

CUDA

Nsight

Triton

Job description

Jane Street seeks compiler engineers to join a new ML Compilers Team in London. You will help translate research ideas into high-throughput training and low-latency inference, shaping the compiler stack and APIs used across ML workloads.

The role involves collaborating with researchers and engineers to optimize graph captures, DSLs, and kernel execution on GPUs, while maintaining clean interfaces and robust tooling.