X.9.3 · Skill in AI-assisted work
Verification
Proving that work, human or AI, actually does what it should.
Verification means running it, testing it and checking it against the requirement. Tests, type checks and real usage provide evidence. Done means verified, not written.
Resources
2- articleStart hereHallucinations in code are the least dangerous form of LLM mistakes ↗Simon Willison · 2025 · 6 min
Errors that fail loudly are easy, while code that runs but is subtly wrong needs real review and testing.
- articleStart hereYour AI product needs evals ↗Hamel Husain · 2024 · 18 min
The practical case for evals, built from a real product, from simple assertions to human review of traces.
Tools
none yetTools for this topic are being researched.