Lead evaluation, selection, and optimization of large language models. Build and maintain eval frameworks, implement RAG and embedding-based pipelines, integrate new models into production on cloud, monitor performance and costs, and own end-to-end deployment, validation, and documentation.