Applied ML Builders

Choose a real problem (LLM eval harness, RAG over your docs, fine-tuning a domain model, anomaly detection). Design the eval. Build the system. Ship it. Present results to a senior engineer panel at week 12.

Focused on practical applicability and tradeoffs — not theory worship. We pick papers you can actually use this quarter and discuss what to ignore from each one.

Monthly real-incident retros. Senior engineers walk through actual outages, eval regressions, and silent quality degradations. Anonymized, gripping, useful.

Battle-tested eval harnesses, dataset versioning patterns, drift detection scripts, and monitoring dashboards. Plug your data in and ship.

Real production patterns for RAG, prompt versioning, structured outputs, fine-tuning, and inference cost optimization. Updated monthly as the field shifts.

Drop in any Friday. Senior engineers from Anthropic, Cohere, and Hugging Face guest-host on rotation. Bring real stuck-points and you'll leave with a path forward.

Open ML roles at member companies, plus mentoring on portfolio building and interview prep. Members regularly land roles up the seniority ladder through this room.