Skip to content

Free The Complete AI Agent Engineering Course

Nitmonk

Blog

Thoughts on AI engineering

Mental models, hard-won lessons and the occasional strong opinion — my notes on how to actually build production AI, written for engineers who ship.

Benchmarks · AI Agents · Jev
Latest

LLMs vs a System One Model: A Chess Marathon Benchmark

I built an arena where frontier LLMs and TypeSafe's Jev play chess move by move — tracking cost, latency and tokens for every move. The real lesson was about honest benchmarking.

8 min read

  • RAGVector Databases

    How Vector Databases Actually Find Your Nearest Neighbors

    Ask a 10-million-document knowledge base a question and it answers in milliseconds. Here is how vector search really works — IVF, PQ, HNSW, DiskANN — and which database to pick.

    9 min read
  • MindsetAI Agents

    The Mental Model Every AI Agent Engineer Should Have

    A viewer asked what AI Engineers actually code all day. Here is the mental model I keep coming back to — the seven problems that AI Agent Engineers and AI Architects really solve, for learning, problem solving and interviews.

    4 min read