Jobs

Callosum is seeking a Member of Technical Staff for Inference Performance & Deployment to bridge internal engineering with real-world performance. This role involves designing tooling and methodologies to systematically characterize performance, identify bottlenecks, and drive production-readiness for heterogeneous AI infrastructure. You will run experiments, develop deployment patterns, and work on orchestration software to optimize latency, throughput, and cost. The ideal candidate has experience deploying and benchmarking large model inference in production environments and strong end-to-end performance characterization skills.