Newsletters

LLMs are not great at following commands. They deviate. Frequently. And they’re unreliable at self-verification—they’ll confidently tell you the code works while it’s on fire. The fix isn’t to ask the LLM to verify. It’s to ask it to write a script that verifies. Shift from judgment to artifact.

Instead of asking an LLM “Did this work?” you define verification steps that produce a series of pass/fail artifacts. The agent can’t negotiate with a failing test. It either meets the specification or it doesn’t.

If you put all these things together:- RLHF = training the model to be likable by humans- RLVR = training the model to be accepted by machines- RLVR is more scalable- "Alignment tax" says "likable by humans" makes the model do worse on verifiable tasks

long-horizon agents assume requirements all exist upfront. They are fundamentally against human in the loop. And they don't have true judgment for what humans like. That—is why I don't like them.