The point where experts and best companies meet
Share
Improve all tooling and automation in use in the team, from simple data collection scripts to datacenter-scale ML CI/CD systems.
Understand and internalize workflows for GPU performance analysis and optimization so you can help us re-invent them.
Build Python-based machinery hooking into common Deep Learning software like PyTorch or JAX to support performance analysis work.
Ruthlessly discover and chase down workflow- and tool-related inefficiencies in the team's daily work, and dream up and implement ways to eliminate them.
MS degree in CS or adjacent fields or equivalent experience
3+ years of relevant work experience
Background in deep learning fundamentals and common deep learning software, especially PyTorch/JAX
Experience in GPU computing, i.e. fundamental understanding of heterogeneous multi-node accelerated computing systems
Background in analyzing and optimizing application performance
Familiarity with containerized CI/CD flows, e.g. gitlab + docker
Programming skills in C++, Python, and CUDA
Deep passion related to tools, scripts, and automation
These jobs might be a good fit