Skip to lesson
Exit
Deep Dive: RL for Coding Agents1 / 2

1 min lesson

Coding is the interesting middle

Match each case in "Coding is the interesting middle" to the signal and response that fit it.

Step 1 of 2

Coding is the interesting middle

A patch either passes the tests or it doesn't, so part of coding reward is purely verifiable. But a passing patch can still be unreadable, slow, insecure or a hack that special-cases the test. Those qualities aren't captured by any test you wrote, which is exactly why Cursor invests in learned graders.

Verifiable slice

Compiles, type-checks, tests pass.

Cheap, deterministic, trustworthy reward.

But: many real edits have weak or no tests.

Non-verifiable slice

Code quality, style, partial credit, multi-file coherence.

No checker exists - needs a learned grader.

This is where Cursor's research charter actually lives.

A learned grader is a reward model specialized to coding judgment - partial credit on a half-right diff, whether a multi-file change hangs together, whether the fix matches the request. Building and trusting these graders is one of the role's named responsibilities.