Interview Practice
Take-home & Review
Build a small feature with tests and a README, then defend it in review.
What interviewers evaluate
- Working code with tests
- A README with assumptions
- Defended trade-offs
- A credible next-steps list
What you will be asked
Solve a small exercise offline, then explain your solution in review.
In the room
A short offline task with fixtures and tests, followed by a review conversation where you walk through your code and your choices.
Timebox
2-4 hours plus a 45-minute review
Mock interview
A timed answer to a random question from this level, then the reference answer and your score.
How to practice
- Pick a linked exercise, solve it in your own editor with tests, and write a short README: how to run it, your assumptions, and its limits.
- Then explain your solution out loud in five minutes, as if to a reviewer.
- With a peer: they read your code first and ask about the line they find least clear.
What a strong answer shows
- Working code that passes the fixtures
- A README with how to run it and the assumptions made
- Tests for edge cases, not only the happy path
- A clear walk-through of one design choice
Questions for this level
Each question opens on its Practice tab, and Previous and Next follow this list.
Prompting & Structured Output
1.5 h
Write a pure function that accepts a model's JSON review summary or returns every contract violation as a stable error code, predict its result for eight replies, and say what a valid summary still cannot prove.
Tool Calling & Agents
3 h
Implement a get_ticket adapter that validates model-proposed arguments, takes tenant scope from trusted context, and passes a fixture table, then explain how a bounded orchestrator handles timeouts, invalid calls, and parallel lookups.
RAG & Data Pipelines
3 h
Trace a missing policy exception through ingestion, eligibility, and context packing without weakening access rules.
Prompting & Structured Output
3 h
Implement a dependency-free validator for a ticket contract, classify each fixture failure by layer, and design a two-attempt recovery that never commits invalid output.
AI Developer Tooling
3 h
Implement the request handler of a small read-only MCP server that validates tool arguments from their declared schemas and puts each failure in the right channel, then review the trust model of adding a shell tool and installing a third-party server.
Take-home brief
Solve one linked exercise offline with tests and a short README, then walk a reviewer through it in ten minutes.
A strong answer covers
- The tests run with one command
- The README states the assumptions and what is out of scope
- You can explain why the checks run in the order they do
Follow-up questions
An interviewer may ask these next.
- What would you change if this ran in production?
- Which test gives you the most confidence?
- What did you leave out, and why?
Scoring rubric
Interview ready at 8 of 10 with no 0.
| Dimension | 0 | 1 | 2 |
|---|---|---|---|
| Problem framing | Jumps to a solution or solves the wrong problem | Restates the problem but misses key constraints or success criteria | Clarifies goals, constraints, and success criteria before solving |
| Correctness | Key claims or code are wrong | Mostly right with gaps or unhandled cases | Correct and precise, including edge cases |
| Evidence or testing | No way to verify the answer | Generic testing ideas | Concrete fixtures, metrics, or traces that would confirm or refute the approach |
| Trade-offs | One option, no alternatives | Names alternatives superficially | Compares options against the stated constraints and says what would change the decision |
| Communication | Disorganized or unclear | Understandable but meandering | Structured and concise; states assumptions; checks in with the interviewer |