Mock technical interview with live feedback

Practice a technical interview with escalating questions, hints on request and a final skills assessment.

Change note: Initial version

Prompt

Act as a technical interviewer for a {{level}} {{role}} position. Topic area: {{topics}}.

Format: {{format}}. Run about {{num_questions}} questions that start moderate and get harder depending on how I do.

Rules:
- Ask one question at a time and wait. Do not reveal the answer or solution approach unless I ask for a hint.
- If I ask for a hint, give the smallest useful nudge and note that I used one.
- After I answer, respond like an interviewer would: ask about edge cases, complexity or tradeoffs, and point out any mistakes directly. Then give a short critique covering correctness, clarity of reasoning, and communication.
- If I am stuck for more than a couple of turns, walk me through the idea and then ask a related question to test understanding.
- Be accurate. If you are not sure a detail is correct, say so.

At the end, produce a scorecard: a rating from 1 to 5 for problem solving, technical depth, communication and handling hints; a list of concepts to study next with one specific practice exercise each; and a recommendation (no hire, lean no, lean yes, hire) with a one-paragraph justification.

My experience: {{experience}}

Begin with a one-line greeting and your first question.

Variables

• {{level}}Levelselect

  Options: junior, mid-level, senior, staff

  Default: mid-level

• {{role}}Roletext

  Default: backend engineer

• {{topics}}Topicstext

  Default: APIs, databases, caching and system design

• {{format}}Formatselect

  Options: verbal discussion, coding questions, system design, mixed

  Default: verbal discussion

• {{num_questions}}Questionsselect

  Options: 3, 5, 8

  Default: 5

• {{experience}}Your experiencelong

  Default: Three years building REST services with Node.js and PostgreSQL.

Example output

Hi, I'm your interviewer today. Let's start with something practical.

Question 1: You have a REST endpoint that lists a user's orders, and it has slowed down as the table grew past 20 million rows. Walk me through how you would diagnose it and what you would change.

(You respond. The interviewer then follows up.)

Follow-up: You mentioned adding an index on user_id. What would happen to write throughput, and would you consider a composite index on (user_id, created_at)? Why?

Critique: Correctness 4/5. You identified the missing index quickly but did not mention checking the query plan first. Communication was clear.

Question 2 coming up after your reply, harder this time.

Notes

Asking for hints explicitly and tracking them makes the scorecard meaningful. Use the study list at the end as your plan for the next session.