AI Platform · MLOps · ML Infrastructure · Production AI
I build production AI and distributed systems with a focus on reliability, evaluation, observability, and controlled execution.
3+ years of production experience across ML lifecycle management, Generative AI, model serving, data platforms, and batch / real-time systems.
- ML Platform Infrastructure — Kubernetes, MLflow, GitOps, autoscaling, observability · 645K+ requests, 0% errors
- Agentic SRE — evidence-grounded incident investigation over metrics, logs, traces, and Kubernetes
- Agentic AI Control Plane — policy-controlled agents, RAG, HITL, safety evaluation · 0 unsafe executions / 540 attempts
- Knowledge Base RAG — hybrid retrieval, reranking, grounding, multilingual evaluation · 95.9% evidence recall
- DecisionSQL — governed Text-to-SQL with deterministic execution boundaries · 88.9% task success
- CauseTune — QLoRA fine-tuning and controlled evaluation · 65.3% → 99.3% exact match
Stack: Python · Go · FastAPI · PostgreSQL · Kafka · Docker · Kubernetes · MLflow · Airflow · dbt · LangGraph · Qdrant · OpenTelemetry
Selected repositories are pinned below ↓
