"Tell me about a time you had to manage out a low-performing engineer. How did you handle the conversation, and what did you do before and after it? If relevant, you can use an example from a team working on Databricks infrastructure, Unity Catalog, or a core platform area where execution quality and trust were especially important."
This question tests whether you can handle one of the hardest parts of engineering management: balancing empathy for an individual with accountability to the team and the business. Interviewers want to see that you did not jump straight to termination, that you diagnosed the performance issue clearly, set expectations, documented a fair process, and communicated directly and respectfully.
It also probes judgment under ambiguity. In many cases, poor performance can stem from role mismatch, unclear expectations, personal circumstances, or team dysfunction. A strong manager can separate those factors, protect delivery on critical work, and still treat the engineer with dignity.
A strong answer is specific: what the performance gap was, how long it persisted, what support you provided, how you structured the conversation, and what outcome followed. The best responses use a clear STAR arc, include measurable impact on the team, and show both compassion and ownership rather than blame or avoidance.