Description
Crusoe is hiring an LLM Inference Engineer to own the end-to-end inference stack in production, improving large language model performance, latency, throughput, and cost. The role involves optimizing serving architectures, profiling and tuning frameworks such as vLLM and SGLang, working down to CUDA kernels, adapting optimization methods across models, and partnering with customer engineering teams to move workloads from proof of concept to monitored production services. The position requires a degree in a relevant field, production coding experience, familiarity with LLM inference optimization and GPU behavior, and strong communication skills; Python is preferred. Compensation is up to $250,000-$300,000 plus bonus, with equity and comprehensive health, dental, and vision benefits.
