Research Engineer, Infrastructure, Numerics
San Francisco
Workplace: HybridFull timeUSD 350,000 - 475,000 annuallyFunction: Research & Scientific (R&D)Education: bachelorsSkills: ["Collaboration","Initiative","Bias for action","Debugging","Problem-solving"]Design and optimize distributed training infrastructure for large-scale LLMs with a focus on numerics. You’ll implement and evaluate low-precision formats, build kernels and communication primitives for mixed/low-precision arithmetic, and prototype scaling strategies across multi-GPU, multi-node setups. Collaborate with research teams to co-design training recipes, and contribute to internal orchestration and monitoring so thousands of experiments run efficiently, stably, and reproducibly.
Loading
Loading job details...
Preparing the role view and application actions.

