Tell me about a project you're proud of.
This open-ended prompt is an invitation to show your best work on your own terms — pick one project that demonstrates technical depth, clear ownership, and measurable impact relevant to the role you're interviewing for. Do not summarize your entire résumé; go deep on one story.
How to think about it
What the interviewer is actually testing
This question reveals what you value — which project you choose says as much as how you describe it. Interviewers are listening for technical credibility (did you actually build this?), ownership (were you the driver or a passenger?), and judgment (does the impact you’re proud of align with things that matter in the real world?). It also serves as a warm-up to deeper follow-ups, so pick a project you can defend from every angle.
How to structure a strong answer
Pick deliberately. Match the project to the role. Applying for an ML engineering role? Choose a project with production complexity, not a notebook experiment. Data analyst role? Choose one where your analysis changed a decision.
Lead with the impact. Open with the headline result — “I built a real-time fraud scoring system that cut false positives by 30% while holding recall constant” — then walk backward into the STAR structure. Starting with impact immediately establishes why the project matters.
Go specific on the Action. Name the methods, tools, and tradeoffs you chose. Generalities like “I built a model” tell the interviewer nothing. “I chose LightGBM over XGBoost because training time on 50M rows was a constraint, and I validated the decision with a time-series cross-validation split to avoid leakage” tells them you know what you’re doing.
Acknowledge one honest challenge. Projects that went perfectly sound made-up. Name one obstacle you hit and how you resolved it — this adds credibility.
Close with a number. Revenue, latency, error rate, adoption rate, cost avoided. If you can’t recall the exact figure, give a range and say so.