// learn · beginner
Never trust one AI: build with one, check with another
Lee builds with Codex and has Claude check the work, with Hermes holding context across models.
credit: Lee Shand
Lee's way
- Do the work in your main agent (Lee uses Codex).
- Hand the result to a second model and ask it to find mistakes.
- Keep shared context in one place so both models see the same facts.