Interview Practice
AI System Design
Design an AI-powered system with explicit guarantees, evaluation, and operations.
What interviewers evaluate
- Explicit guarantees and data flow
- Security boundaries
- An evaluation plan
- Cost and latency budgets
- Incident recovery
What you will be asked
Sketch the flow of a simple AI feature end to end.
In the room
The interviewer asks you to draw the main components and data flow of a retrieval-grounded assistant or a tool-using assistant, and to name one risk per stage.
Timebox
45-60 minutes
Mock interview
A timed answer to a random question from this level, then the reference answer and your score.
How to practice
- Use the linked questions: draw the ingestion and query paths, or the tool call lifecycle, from memory.
- Label who is trusted at each step: the user, the model, the application, and the tool.
- With a peer: they point at one arrow and ask what happens when it fails.
What a strong answer shows
- Names each stage in order and what it produces
- Shows where access control is enforced
- Names one failure per stage and how the user sees it
- Keeps the design small enough to explain in ten minutes
Questions for this level
Each question opens on its Practice tab, and Previous and Next follow this list.
RAG & Data Pipelines
1.5 h
Explain how documents become evidence for a generated answer, and why an answer can be wrong even when it cites a real document.
Tool Calling & Agents
1.5 h
Follow one ticket lookup from tool declaration to answer: what the model proposes, what the application validates and executes, and what the assistant says when the result is unavailable.
AI Developer Tooling
1.5 h
Explain how a coding assistant assembles its context, which component decides whether each proposed action runs, and why an instruction found inside a file cannot grant itself permission.
Follow-up questions
An interviewer may ask these next.
- Where does the tenant come from on each request?
- What happens when a source document is updated?
- How would you know the assistant is giving worse answers?
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 |