Refreshing or reusing an interview URL allows restarting a known problem

[ZuoTso]

Description

While using CodeTrial for interview preparation, I noticed that the interview URL contains the selected problem explicitly, for example:

/interview?problem=kiosk-panel-mirror-audit&duration=45&loop=coding_behavioral

Refreshing or reopening this URL allows the same known problem to be started again.

I am not sure whether this is intentional behavior for practice, reproducibility, or development purposes. However, in an interview-simulation context, it also makes it easy to restart after already seeing the problem, which may reduce the usefulness of the session as a realistic self-assessment.

Questions / expected behavior

Would it make sense for an interview session to keep its existing state across refreshes, rather than starting the same problem again from the beginning?

Also, is the problem query parameter intentionally user-selectable as part of the public interface, or is it mainly intended for internal/practice use?

If this behavior is intentional, documenting the distinction between practice/reproducible sessions and stricter interview simulation may be helpful.

[jserv]

I am not sure whether this is intentional behavior for practice, reproducibility, or development purposes. However, in an interview-simulation context, it also makes it easy to restart after already seeing the problem, which may reduce the usefulness of the session as a realistic self-assessment.

Make it better by submitting your pull request(s).

[ZuoTso]

Thanks. I haven't contributed to this project before, so I'll first look into the relevant code and try to understand the current session/refresh behavior. If I can isolate a small change, I'll prepare a PR.

[jserv]

I haven't contributed to this project before, so I'll first look into the relevant code and try to understand the current session/refresh behavior. If I can isolate a small change, I'll prepare a PR.

Avoid meaningless statements like the one above. In the era of AI, "firsts" are everywhere: the first time using a frontier AI model, the first collaboration, and so on. Focus on constructive discussion and useful information instead.