Specializes in retrieval-augmented generation architecture — building the search, indexing, and retrieval layer that grounds LLM outputs in accurate, up-to-date information.
Works on pretraining and post-training of foundation models — architecture experiments, training efficiency, alignment techniques (RLHF, DPO), and scaling research, often in collaboration with research scientists.
Owns model development, training, and quality — building and improving the ML models (not just LLM wrappers) that power a product, from data pipeline to production deployment.
Owns the compute infrastructure that trains and serves AI models — GPU cluster management, distributed training pipelines, and inference serving at scale. A natural pivot for DevOps/backend infra engineers.
