Re-Solving Problems You Have Already Done vs Fresh Unseen Problems in Timed Mocks
At some point most candidates look at their solved list and wonder whether to go back. Re-solving old problems is appealing: you already know roughly what to do, it goes faster, and each clean solve feels like proof the pattern stuck. The other path is fresh, unseen problems as timed mocks—a problem you have never read, a real-length timer, the plan said out loud, and a dry run before "done."
Both are honest practice, and good prep uses both. They train different things. The useful question is: when does repetition help you in a live round, and when does it just feel like progress?
This is a workflow comparison, not a claim about how memory works for everyone. People retain patterns differently, and some problems are worth repeating many times. What stays the same is that a live round almost always gives you a problem you have not seen in that exact form.
At a glance
| Category | Re-solving problems you have done | Fresh unseen timed mocks (Luna side panel optional) |
|---|---|---|
| Time per problem | Usually shorter | Matches your real round length |
| What it trains most | Pattern recall, clean implementation | Recognizing patterns cold, getting unstuck |
| How it feels | Smooth and confident | Uncomfortable, sometimes slow |
| Risk | Remembering the answer, not the idea | Fewer clean solves, more frustration |
| Best for | Locking in core patterns | Rehearsing the live round |
| Main tradeoff | Little practice at starting from zero | Patterns can fade without review |
What re-solving old problems looks like
A typical session:
- You pick problems you solved weeks ago, often from one topic.
- You recognize most of them within a minute.
- You write the solution, usually faster than the first time.
- A few you forgot; you look back at your old answer.
- You mark them done again and feel more confident.
Honest strengths. Repetition helps patterns stick. Re-solving a classic problem—a sliding window, a breadth-first search, a basic dynamic programming table—until you can write it cleanly and quickly frees up attention in the interview for the parts that are new. It is a good way to find patterns you thought you knew but did not. It is also a confidence boost after a bad week, and it is efficient for reviewing a weak topic before an interview.
Honest limits. When you remember the answer, you skip the hardest part of a real round: figuring out which approach fits a problem you have never seen. A smooth re-solve can feel like readiness when you are mostly recalling. It is easy to re-solve the same comfortable set many times and avoid the uncomfortable ones. And because you already know where it goes, you rarely practice clarifying questions, explaining a plan, or recovering from a wrong start.
When re-solving wins: you are reviewing core patterns you will need again; you missed a problem the first time and want to check you really understand it now; a topic feels shaky; or you need a confidence session after a rough mock.
What fresh unseen timed mocks look like
The fresh version keeps the problem new and adds live-round conditions.
- Pick one problem you have never read, ideally from a mixed list without topic labels.
- Set a timer to your real round length, often thirty-five to forty-five minutes including talking.
- Restate the problem out loud and ask the questions you would ask an interviewer.
- Say a two-line plan and the expected complexity before writing code.
- Narrate while you code, and dry-run three cases before you say "done."
- Write one line on where the time went and one thing to change.
Optional companion: Luna Interview is a Chrome extension framed as a practice / calm side panel—not a substitute for knowing the material, and never an "undetectable" live cheat. Keep it beside the editor during the mock while you own the plan, the code, and the testing.
Honest strengths. This is what the interview actually asks of you: see something new and find a way in. You practice the first few minutes, where many people freeze. You learn which patterns you can recognize cold and which only feel familiar after you have read the answer. Talking and testing get practiced every session. And a mixed list keeps you from relying on the topic label to tell you the approach.
Honest tradeoffs. It is slower and more uncomfortable, and many sessions will not end with a clean solution. Without review, patterns you learned weeks ago can fade. It needs a steady supply of problems you truly have not seen. And on a bad day, a string of hard fresh problems can knock your confidence.
When fresh timed mocks win: your interview is close; you do well on familiar problems but struggle on new ones; you freeze at the start of live rounds; or your re-solves have started to feel like recitation.
Decision table for candidate situations
| Your situation | Prefer re-solving | Prefer fresh timed mocks |
|---|---|---|
| Core pattern still feels shaky | Strong fit | Light |
| Interview in the next two weeks | A short review set | Primary lane |
| Strong on familiar problems, weak on new ones | Light | Strong fit |
| Freeze in the first few minutes | Light | Strong fit |
| Missed a problem and read the solution | Re-solve in a few days | Then try a similar fresh one |
| Confidence low after a bad mock | One or two easy re-solves | Return to fresh next session |
| Day before the interview | One or two familiar warm-ups | One easy fresh problem, timed |
A fair practical method (combine both in one week)
Re-solve on purpose, then test the pattern on something new.
Fresh timed mocks (three a week): One unseen problem from a mixed list as a timed mock with restate, two-line plan, narration, and three dry-run cases before "done." Keep a calm practice side panel beside the editor if you use Luna. Track two things only: did you find an approach without help, and did you test before saying "done."
Targeted re-solve set (one short session): Re-solve only the problems you missed or needed hints on in the past two weeks, without looking at your old answer. Explain each out loud in a sentence or two before coding.
Transfer check: After a re-solve, try one fresh problem that uses the same idea but looks different. If you can solve that one, the pattern has stuck.
Common mistakes / when NOT to use each
Skip re-solving when you are repeating the same comfortable problems, when you can type the answer before you could explain it, or when it has replaced fresh practice in the weeks before an interview.
Skip timed-mock theater when you pick problems you have secretly seen, pause the timer whenever you get stuck, or work in silence instead of narrating. Also skip any practice-tool framing that promises stealth help in a real employer interview—that is the wrong goal and the wrong integrity bar.
For both: solutions and AI are for study and postmortems, not mid-session rescue if you are training transfer. The skill is solving a problem you have never seen, out loud, in the time you will actually have.
Soft close
If your bottleneck is getting started on problems you have never seen, Luna Interview is a Chrome side panel built for that practice loop, not for cheating or replacing study. Soft trial path: https://lunainterview.xyz/
Re-solve to remember. Practice fresh to perform. The interview problem is new, so make sure most of your timed practice is too.
Rehearse with an AI interview copilot, then browse commonly reported interview questions.