← All roles

Machine Learning Performance Engineer

Jane Street· Trading, Research, and Machine Learning
Apply on company site
Hybrid (Hong Kong)full-timePosted 3mo ago

About the role

Engineer with experience in low-level systems programming and optimisation to join Jane Street's ML team. Focus on optimizing performance of ML models for training and inference in a high-frequency trading environment, with a whole-systems approach including storage, networking, and GPU-level considerations.

Responsibilities

  • optimising the performance of our models – both training and inference
  • improving efficient large-scale training, low-latency inference in real-time systems and high-throughput inference in research
  • developing and applying a whole-systems approach (storage systems, networking, host- and GPU-level considerations) to achieve goodput in addition to throughput
  • analyzing and optimizing CUDA-related performance at both high and low levels
  • collaborating with ML and infrastructure teams to ensure scalable, efficient GPU training across clusters

Qualifications

  • Understanding of modern ML techniques and toolsets
  • Experience and systems knowledge to debug a training run’s performance end to end
  • Low-level GPU knowledge of PTX, SASS, warps, cooperative groups, Tensor Cores and memory hierarchy
  • Debugging and optimisation experience using CUDA GDB, NSight Systems, NSight Compute (nsight-systems and nsight-compute)
  • Library knowledge of Triton, CUTLASS, CUB, Thrust, cuDNN and cuBLAS
  • Intuition about latency and throughput characteristics of CUDA graph launch, tensor core arithmetic, warp-level synchronization and asynchronous memory loads
  • Background in Infiniband, RoCE, GPUDirect, PXN, rail optimisation and NVLink, and how to use these networking technologies to link GPU clusters
  • Understanding of collective algorithms supporting distributed GPU training in NCCL or MPI
  • Inventive mindset with willingness to question approaches and tooling
  • Fluent in English

Skills

Required:Modern ML techniques and toolsetsEnd-to-end performance debugging of training runsPTX, SASS, warps, cooperative groups, Tensor CoresCUDA debugging/profiling tools (CUDA GDB, NSight Systems, NSight Compute)CUDA libraries (Triton, CUTLASS, CUB, Thrust, cuDNN, cuBLAS)Networking for GPU clusters (Infiniband, RoCE, NVLink, GPUDirect, PXN)Distributed GPU training (NCCL or MPI)English fluency

Benefits

  • Jane Street is an Equal Opportunity Employer
Last updated Oct 25, 2026. Interested? Apply on the company site.

About Jane Street

SIX

Jane Street is a quantitative trading firm and liquidity provider with a unique focus on technology and collaborative problem solving.

Machine learningquantitative tradingtechnologyresearch-drivenglobal marketsinternshipseventsprograms
Company profileWebsite

More open roles at Jane Street