Lead Machine Learning Engineer, Inference & Performance

Configure autoscaling for serving; Deploy LLM models in production; Handle mixed workloads gracefully; Identify training bottlenecks; Implement batching and quantization; Improve throughput per dollar; Manage KV cache strategies; Measure GPU utilization; Minimize latency; Operate models on GPU clusters; Optimize LLM inference; Optimize attention implementations; Profile training runs; Translate client requirements into AI architectures; Tune LLM serving throughput;