Use when identifying code smells and their fixes — smell→refactor mapping (Long Function→Extract, Primitive Obsession→Value Object, Duplication→Parameterize), with detection signals and worked examples.

Use when committing code — one logical change per commit, Conventional Commit format, body explains WHY not WHAT, and clean splitting of mixed changes.

Write complete, runnable code with no TODO, ..., pass, or placeholder markers — every path is filled in. Use whenever generating code so you never hand back a skeleton.

Estimate task size and flag unknowns before committing to a plan — T-shirt sizing, dependency mapping, and surfacing the riskiest unknown. Use when asked "how long will this take."

Decide what belongs in the agent's live context vs what to offload to files/notes so long tasks don't degrade. Use when a session is getting long or context is filling up.

Guarantee valid structured output at the token level — XGrammar, Outlines, GBNF, and provider strict APIs — with the format-vs-semantics distinction that saves projects.

Curate the finite context window of an AI agent — compaction, tool-output clearing, and the principle of finding the smallest set of high-signal tokens — for long-horizon reliability.

Fine-tune an embedding model for your domain — contrastive training with sentence-transformers, hard-negative mining, and Matryoshka representation learning for truncatable vectors.

Build a high-quality fine-tuning dataset — near-deduplication (MinHash/semantic), quality filtering, benchmark decontamination, and data mixing — so your model learns signal, not memorized noise.

Distill a large teacher model into a smaller, cheaper student — logit distillation, synthetic-data distillation, and the scaling laws that decide when it's worth it.