Skip to main content

GPU Kernel and ML Systems Researcher (Research Coding Tasks)

Technology
afterquery
3 days ago
Full time

Job description

Paid, remote contract work with AfterQuery, building and reviewing technical tasks that help test AI models in ML systems and GPU kernels. For researchers and research engineers who write and ship research code.

Domain: ML Systems and GPU Kernels (CUDA, Triton and CUTLASS kernels, inference serving and ML compilers) Requirements

Experience: hands-on research coding in ML systems and GPU kernels

Preferred

Published research with released code in ML systems and GPU kernels PhD in CS or a related field, completed or in progress The work Hands-on research coding in your own area, close to the work you already do. Each task is self-contained, and the project lead walks you through the full scope and a worked example once you're onboarded. Work is asynchronous, so you can fit it around your own research.

Pay $50/hr. Flexible, independent-contractor work with no fixed weekly schedule.

Application process Candidates who meet the requirements and provide the requested materials will be prioritised for review. As part of this process we conduct thorough background checks. Please apply only if you meet these requirements.

Responsibilities

  • Build technical research coding tasks in ML systems and GPU kernels
  • Review and give feedback on AI model outputs in your area
  • Work with the project lead to keep tasks at research quality

Required Qualifications

  • Hands-on experience writing research code in ML systems and GPU kernels
  • Able to write clearly about technical work in your field
  • Able to work independently and manage your own time. This is flexible work with no fixed weekly schedule

Preferred Qualifications

  • Published research at venues such as MLSys, OSDI, SOSP, PPoPP or SC, or comparable
  • Released research code (GitHub or artifact link)
  • PhD in computer science or a related field, completed or in progress
  • First-author publications in your area
  • Comfortable with Docker and reproducible research environments
  • Depth in CUDA, Triton and CUTLASS kernels, inference serving and ML compilers

Why apply

  • Hands-on research coding in your own field
  • Fully remote, asynchronous and flexible
  • $50/hr, paid through Stripe

Interested in this role?