Description
Radical Numerics is hiring a Member of Technical Staff, Inference to build and optimize production systems for large-scale genome and multimodal biological AI models. The role focuses on improving inference latency and throughput, developing GPU kernels and numerical operators, deploying foundation models across cloud, enterprise, and secure environments, and partnering with customers and research teams. Candidates need production experience with large language or foundation models, strong performance engineering and kernel development skills, GPU and CUDA or Triton expertise, Python and PyTorch proficiency, and systems-level troubleshooting ability.
