Engineering Blog • LinkedIn • X (Twitter)
Building boring, reliable infrastructure for exciting AI problems.
I bridge the gap between "it works in a notebook" and "it works for 200 million users." My career has evolved from handling massive data scale at Fandom to architecting MLOps platforms at Qwak, and now modernizing ML infrastructure for high-performance training.
Status Update: I am currently transitioning back to full-time engineering leadership.
Actively looking for Principal / Staff AI Engineer roles (Remote or Hybrid (EU)).
- Production-Grade GenAI: I specialize in Deterministic RAG and Agentic workflows. I move beyond prompt engineering to schema enforcement (BAML), automated evaluation (MRR/Faithfulness), and hallucination mitigation.
- Data Platforms at Scale: Architecting distributed systems (Spark, Ray) and low-latency serving layers that handle billions of events.
- MLOps Infrastructure: "Shift Left" data quality, automated training pipelines on Cloud/K8s, and Cost-Aware Engineering (FinOps).
- Languages: Python, Scala, SQL
- GenAI: LangChain, LlamaIndex, BAML, Vector Databases (Chroma/Qdrant/Milvus)
- Compute & Data: Apache Spark, Ray, Kafka, Airflow, DBT
- Infrastructure: AWS, Kubernetes, Terraform, Docker, Istio
- O'Reilly Contributor: Co-Author of "97 Things Every Data Engineer Should Know" (Chapter 76).
- Speaker: Berlin Buzzwords, Data Natives, Infoshare, LambdaDays.
- Educator: Trained 1,000+ engineers in AI/ML architectures.