AI Engineer with 5+ years of engineering experience and an M.S. in Artificial Intelligence, specializing in production LLM and RAG systems. Shipped a multi-tenant Text-to-SQL + RAG analytics platform over 50M+ POS rows at a 90%+ benchmark pass rate, built an LLM evaluation pipeline (LLM-as-judge, RAGAS), and cut LLM inference cost by 65%.
Built an end-to-end MLOps platform — versioned data pipelines, MLflow/Optuna tracking, CI/CD promotion gates, BentoML serving, and drift-triggered retraining — config-driven via ~30 YAML registries.