prompt optimization, caching strategies, batching, token efficiency, model selection for latency trade-offs. Measure and optimize end-to-end latency for agent workflows, including LLM inference, orchestration, tool invocation, and response generation. Performance culture and team leadership Build and lead the Performance Engineering team, including specialists in frontend performance, backend … improving performance. Deep expertise in specific performance domains: edge computing, database optimization, distributed systems optimization, microservices performance. Experience optimizing AI/ML inference pipelines, LLM serving, or agentic system performance at scale. Experience working in fast-growing organizations where product velocity and operational efficiency are critical competitive advantages. Published content ...