Senior LLM Inference Engineer / AI Infrastructure Architect - 20159
Core
Analyze and optimize LLM-powered applications and agent workflows for performance, cost, and capacity.
Role type
Senior IC AI Infrastructure Architect / LLM Inference Engineer
Builds
Optimized LLM inference services and agent workflows
Domain
AI/ML Infrastructure, Large Language Models
Deliverable
production ML models
Required skills
LLM inference optimization, GPU memory management, Python, Linux, container deployments, vLLM, TensorRT-LLM, load testing, capacity planning, quantization, batching, prompt caching
Preferred skills
AI agents, distributed inference, multi-GPU scaling, cloud/self-hosted/hybrid operating models
Technologies
vLLM, TensorRT-LLM, Python, Linux, GPU
Responsibilities
Analyze LLM applications to identify bottlenecks and cost drivers; Optimize inference performance regarding token usage, latency, and concurrency; Evaluate model selection for task complexity; Implement optimization techniques like quantization and batching; Conduct load testing and capacity planning; Compare operating models (API, self-hosted, on-prem, hybrid); Develop strategic recommendations for GPU capacity and scaling; Document findings and transfer knowledge to the team.