From the desk
I build and operate production AI systems — document intelligence pipelines, retrieval on vector stores, and LLM serving on Kubernetes — the parts that decide whether a clever demo survives contact with real documents and real users.
This is where I write it down: long essays on what actually broke, course series that go deep on one thing at a time, and the interview notes I wish I had while preparing.
† Interview preparation
Notes filed by topic
Everything I revise from — data structures, machine learning, RAG, MLOps, system design and the rest — organised so you can open one folder and stay there.