Solving Out Loud

Beginner · 14 min read · Interview Playbook, Interview Playbook

How DSA interviews are scored and what to say at each step.

Many candidates who could solve the problem still fail the interview, and many who don't finish still pass. The difference is almost always communication. The interviewer can't see inside your head. All they can score is what you say and write.

What the interviewer is scoring

Most companies use a rubric close to this one:

The 45-minute shape

Phase 1: clarify (0–5 min)

Repeat the problem in your own words, then ask about the things that change the algorithm:

Phase 2: approach (5–15 min)

Say the brute force out loud first, even if you already see something better. It proves you understand the problem and gives the interviewer something to anchor on.

Phase 3: code (15–35 min)

What "good code" means to an interviewer, in the order they notice it:

Phase 4: test (35–42 min)

Don't announce "I think it works." Show it:

Phase 5: wrap up (42–45 min)

State the complexity with the reason:

When you're stuck

Being stuck happens to everyone in interviews. What matters is what you do next. Work through this list out loud:

If you've seen the problem before

Say so: "I've seen this one, I think it's solved with a monotonic stack." Interviewers ask questions to watch you reason. If you recite a memorised answer and they find out, it reads as dishonest. If you're upfront, it earns trust, and they'll either give you another problem or ask you to explain it properly anyway.

Phrases that help

Interview questions

  • What should you do if you realise halfway through coding that your approach is wrong?
  • Is it bad to ask the interviewer for a hint?

Further reading

Key takeaways

  • Interviewers score problem solving, coding, verification, communication and complexity, not just the final answer.
  • Clarify, approach, code, test, wrap up: roughly 5, 10, 20, 7 and 3 minutes.
  • Always say the brute force, then name its waste and the fix.
  • Test unprompted: trace an example, then edge cases. Fixing your own bug is a plus.
  • When stuck, think out loud, use the framework, and ask targeted hint questions.

Previous: A 10-Week Study Plan · Next: Spot-the-Pattern Drills