How would you explain repo grounding to an interviewer?

Example Answer

The way I'd approach it in an interview is this: Repo grounding means the agent is reasoning from the actual codebase, tests, docs, and conventions in front of it instead of hallucinating from generic language patterns. For coding systems, that is the difference between plausible code and defensible code.

A grounded agent knows where the relevant files are, what APIs already exist, what tests or docs constrain the change, and what local patterns should shape the patch. Without that, it may still produce fluent code, but the odds of subtle mismatch go up fast.

In practice, repo grounding is what turns a model from a code generator into a software-engineering assistant.

What I always try to avoid is giving a process answer that sounds clean in theory but falls apart once the data, users, or production constraints get messy.

Common Poor Answer

A weak answer is saying repo grounding just means giving the model the repository. The important part is retrieving and reasoning from the right repo context.