Hirestack
Hirestack
Chapter 22

Career · The AI engineering interview

📖 3 min read · 7 sections · May 2026

Why this matters

You can know all 21 prior chapters cold and still flunk the interview. The interview rewards a specific pattern of communication — structured, confidence-calibrated, tradeoff-aware. This chapter is the meta-skill.

The typical loop (2026)

StageWhat they assessHow long
Resume screenPast production AI work, depth signalspassive
Recruiter callComp, fit, role expectations30 min
Technical screenFoundations: tokenisation, embeddings, RAG basics, transformer mechanics1 hour
CodingBuild something with an LLM — RAG, agent, structured extraction1-2 hours
System designDesign a real LLM product end-to-end1 hour
Behavioural / leadership (senior)Project ownership, cross-team work, ambiguity handling1 hour

Topics by level — what's expected

LevelBar
L4 (mid)Strong at prompts, structured output, basic RAG, can ship one feature
L5 (senior)Owns RAG end-to-end: chunking, eval, observability. Picks the right vector DB, defends the choice. Knows when not to use agents.
L6 (staff)Architects production AI systems: cost, safety, multi-agent orchestration, evaluation strategy, organisational governance.

The system-design interview pattern

Mirrors the HLD system-design method:

  1. Clarify (3 min): scale, latency, accuracy target, budget, what's in/out of scope.
  2. Estimate (3 min): QPS, tokens/request, cost/month, vector store size.
  3. Sketch (10 min): blocks — entry point, retrieval, LLM, output filter, observability. Name the tools you'd use.
  4. Deepen (15-20 min): pick two boxes the interviewer probes; show real tradeoffs. The interviewer will steer here.
  5. Defend under cross-examination: "what if X breaks?", "what does this cost at 10× scale?", "how do you evaluate?".

Common questions with real answers

Q: How do you evaluate a RAG system?
Three layers: (1) Golden set — 100-500 hand-curated Q&A pairs covering easy/medium/hard/edge. (2) RAGAS metrics on every deploy — faithfulness, answer relevance, context precision/recall. (3) Online sampling — 1-2% of production traffic gets LLM-judge eval; track rolling-window scores; alert on 2σ drift. The golden set runs in CI; RAGAS and online go to dashboards.
Q: Hybrid retrieval — explain when it beats dense alone.
Hybrid (BM25 + dense) wins when the data contains exact-match tokens that semantic embeddings underweight: SKU codes, error codes, function names, acronyms, named entities. BM25 catches those; dense catches paraphrased semantic similarity. Combine with reciprocal rank fusion. Pure dense is fine when your data is prose and queries are conceptual.
Q: A user reports your bot is confidently wrong. Walk me through diagnosing.
(1) Pull the trace — what did retrieval return? (2) Were the right chunks in the top-5? If not, retrieval bug — check chunking or query rewriting. (3) If retrieval was right but model still wrong, check the prompt: was "answer only from context" present? Were chunks clearly delimited? (4) If prompt was right but model still wrong, try Sonnet/GPT-4o instead of the cheaper tier — sometimes the model just isn't capable enough. (5) Add this case to the golden eval set so the next regression catches it.

Behavioural questions for senior roles

Red flags interviewers watch for