12+ years building production AI/ML systems at scale. 8 years at Intuit shipping tax intelligence, CTO of an internal venture (0-to-GA in 9 months), and identity migration across 100+ microservices. Open to Principal/Staff+ AI Engineering roles in Canada.
Subscribe to my newsletter for insights on production AI systems, engineering leadership, and practical strategies for building reliable AI at scale.
- Production-Scale ML Systems: governed execution layers, identity systems.
- Agent Reliability Engineering: Multi-agent orchestration with ReAct loops, token budgeting, HITL gates, and failure taxonomies
- RAG & Retrieval Systems: Permission-aware retrieval (pgvector + RLS), hybrid BM25 + dense with RRF, NLI entailment verification, RAPTOR synthesis
- LLMOps & Observability: OTel GenAI conventions, LLM-as-judge evaluation harnesses, multi-tenant pipelines, A/B testing infrastructure
- 0-to-1 Technical Leadership: Owned architecture, hiring, and delivery as CTO — comfortable driving ambiguous problems from whiteboard to production
- Multi-Agent Orchestration & LangGraph
- RAG Pipelines & Vector Databases (pgvector)
- LLM Evaluation, Calibration & Fine-Tuning (LoRA)
- Distributed Systems & Temporal
- OpenTelemetry & LLMOps
- Mechanistic Interpretability (logit lens, published research)
- Python, Production ML at Scale