AI Engineering Notes
Notes on LLM systems, agents, evaluation, and the patient work of turning fast-moving AI ideas into reliable software.
Current Focus
Architecture notes on retrieval, tool use, memory, context, and orchestration.
Practical experiments with agent loops, workflows, observability, and failure modes.
Methods for making model behavior visible, measurable, and less dependent on vibes.
Latest
A researched essay on why AI coding work is shifting from one-shot prompts toward feedback loops, harnesses, verification, and agent operating systems.
A first note on building a public learning practice around LLM systems, agents, evaluation, and the engineering habits behind them.