Skip to lesson
Exit
Capstone: Mock Loop & Self-Exam1 / 2

1 min lesson

Worked sketch - Drill 1, the non-verifiable grader

Design an online-RL setup for a Tab-style feature. What are the four pieces you must address and the chief risk of training on live traffic?

Step 1 of 2

Worked sketch - Drill 1, the non-verifiable grader

How to answer it
Signal
Blend a learned reward model with cheap verifiable checks (compiles, lints, partial test pass) so it isn't a single gameable scalar.
Partial credit
Score multi-file edits per-hunk against intent, not all-or-nothing, so near-misses get gradient.
Anti-gaming
Hold an adversarial set of reward-hacked stubs; require the grader to rank them below real fixes or it's rejected.
Drift guard
Retrain the reward model on fresh policy samples and keep a KL leash so the policy can't run off into the grader's blind spots.
Interview move

Lead every design answer with the failure mode, then the mitigation. “The obvious risk is the policy learns to satisfy the grader's proxy instead of the task, so I'd hold an adversarial set and require the grader to rank hacks below real fixes.” Anticipating the hack unprompted is the single clearest signal of a researcher who has actually trained these systems, not just read about them.

Don't design in a vacuum, ground it in their work

Generic RL design reads like a textbook. Tie the online-RL drill to the published Tab approach and the grader drill to non-verifiable coding rewards and reference the constraints they actually face - async multi-region pipelines, sandboxed environments, training on real sessions. Grounded specifics separate a candidate who studied Cursor from one who studied RL.