Inference Optimization Engineer

This role focuses on optimizing LLM and VLM inference serving backends, evaluating engines, and resolving production regressions to improve latency, throughput, and cost efficiency. The engineer will deploy and extend modern inference technologies, build model-compression workflows, and implement advanced inference techniques to enhance performance. Collaboration with kernel and platform engineers is essential to identify bottlenecks and guide optimization decisions through quantitative analysis and benchmarking.