Proudest Project, Using AI, Handling a Disagreeing Customer, and Why Change Roles

In a recruiter screen for a data scientist role, most of the conversation is behavioral. Answer the four questions below as you would on that call, using specific examples from your own work. ### Clarifying Questions - Should the proudest project come from a data science role, or can it be any project, including one outside work? - For the AI question, does the recruiter mean the AI tools you use day to day, the AI or machine learning you build into products, or both? - In the customer question, is the customer an external client or user, or an internal stakeholder, such as a product or business partner, who consumes your analysis? - Does the recruiter want one example per question, or is a general approach illustrated by an example acceptable? ### Part 1 — Your proudest project Tell me about the project you are most proud of. ```hint Pick for this listener A recruiter may not follow the technical details. Choose a project whose impact you can explain in one plain sentence, and be clear about which parts you did yourself. ``` #### What This Part Should Cover - Why this project was chosen, and the problem and stakes in plain language - The candidate's own decisions and actions, separated from the team's - The outcome, measured where possible, and what made the project hard ### Part 2 — How you use AI How do you use AI in your work? ```hint Beyond the tool list Naming tools is the easy part. Think about where you check what the tool produces, and where you would not rely on it. ``` #### What This Part Should Cover - Concrete uses inside a data science workflow, not a list of product names - How outputs are verified and how sensitive data is protected - The effect on speed or quality, and the limits on where AI is used ### Part 3 — A customer with an opposing view How do you deal with a customer who holds a view opposite to yours? ```hint Locate the disagreement Ask whether you disagree about the facts, the method, or the goal. Each one is resolved differently. ``` #### What This Part Should Cover - Understanding the customer's position and the constraints behind it before arguing - Moving the disagreement onto evidence or an agreed decision criterion - How the decision was made and the relationship afterward, including what happens if the candidate turns out to be wrong or is overruled ### Part 4 — Why you are looking for a new role Why are you looking for a new role? ```hint Toward or away Is your answer mostly about what you are leaving or about what you want next? Consider which one tells the recruiter more about whether this role fits you. ``` #### What This Part Should Cover - An honest motivation that does not criticize the current employer, team or manager - A clear link between what the candidate wants next and what this role offers - Consistency with the career story told in the other answers ### What a Strong Answer Covers - Specific first-person examples with actions and results, not general statements about how the candidate usually works - Answers sized for a screening call, with technical detail translated into business impact - A consistent story across the four answers, so the proudest project, AI habits, way of handling disagreement and reason for moving point to the same strengths - Judgment and candor: trade-offs, mistakes and limits acknowledged, with no blame placed on customers or colleagues ### Follow-up Questions - If you started your proudest project again today, what would you do differently? - Tell me about a time an AI tool gave you a wrong or misleading result. How did you catch it, and what did you change afterward? - Suppose the customer turned out to be right and your analysis was wrong. How did you find out, and what did you do? - What would your current team need to offer for you to stay?