ML Compiler Engineer - Drive High-Throughput ML Stacks

Jane Street Group; New York (NY); On-site; USD 270,000 - 330,000; Full time

Job summary

Jane Street is hiring a compiler engineer to join our ML Compilers Team in New York. You will shape and extend the compiler stack to turn research ideas into high-throughput training and low-latency inference workflows, collaborating across ML and hardware teams.

The ideal candidate has ML compiler experience, familiarity with PyTorch/Jax/TensorFlow, and knowledge of graph compilers and kernel DSLs. Strong in performance and system design, you’ll help advance our ML tooling and productivity.

Qualifications

• Experience with ML compilers or compiler stacks.
• Familiarity with high level ML frameworks (PyTorch, Jax, TensorFlow).
• Understanding of graph compilers and kernel DSLs (Triton, Helion, cuTile).
• Ability to integrate ML stacks and compiler components.

Responsibilities

• Shape the development of the ML compiler stack.
• Turn research ideas into scalable training and low-latency inference pipelines.
• Collaborate across ML, infra and hardware teams to improve performance.
• Contribute to design and implementation of end-to-end compiler workflows.

Skills

ML compiler

PyTorch

Jax

TensorFlow

Graph compilers

XLA

TVM

Triton

Helion

cuTile

kernel DSLs

Tools

CUDA

CuTeDSL

Job description

Jane Street is hiring a compiler engineer to join our ML Compilers Team in New York. You will shape and extend the compiler stack to turn research ideas into high-throughput training and low-latency inference workflows, collaborating across ML and hardware teams.

The ideal candidate has ML compiler experience, familiarity with PyTorch/Jax/TensorFlow, and knowledge of graph compilers and kernel DSLs. Strong in performance and system design, you’ll help advance our ML tooling and productivity.