Jobs

This role focuses on optimizing and scaling AI infrastructure for foundation model training. The successful candidate will drive down training convergence times by eliminating bottlenecks, designing distributed PyTorch systems for GPU clusters, and implementing efficient low-level code. They will optimize workloads for hardware efficiency and develop monitoring tools for large-scale runs. We are looking for someone with deep experience in distributed systems, ML infrastructure, or HPC, strong Python skills, and mastery of low-level performance optimization with technologies like CUDA, cuDNN, and Triton.