Making Unity legible to coding agents
How I give coding agents enough access to inspect a Unity change, see the rendered result, and repair obvious mistakes before review.
Read field noteField notes
Experiments, with the method left in.
Concise accounts of what I tried, how I checked the result, what failed, and what I would carry into the next project.
A risk-tiered approach to AI adoption
A practical way to vary specifications, review, and access according to the cost of getting an AI-assisted change wrong.
Read field noteA reviewable multimodal production pipeline
How I organise generated 2D, 3D, audio, voice, and video as a traceable production process rather than a collection of one-off prompts.
Read field noteEvaluating open-weight models after fine-tuning
How I compare LoRA adapters for Qwen, Llama, and Mistral against a held-out task set, an explicit baseline, and the realities of serving them.
Read field noteBuilding web performance into everyday delivery
How I keep rendering, caching, third-party code, SEO, and Core Web Vitals inside the normal delivery loop on international commerce platforms.
Read field note