Process: 1. Phone Interview (30 mins) — with the hiring manager Focused heavily on object-oriented design (OOD), with very little on modeling, statistics, or core data science. 2. Take-Home Assignment — anomaly detection No label was provided, so it wasn’t possible to meaningfully evaluate the result. That’s acceptable, but a bit awkward for a DS take-home. 3. Live Technical Interview (Scheduled 90 mins, ended in ~30) This round exposed serious process flaws. Major Issues: • 4 interviewers joined at once. Only one majorly spoke; the others remained silent after brief introductions. • No coding environment was prepared. Midway through, I asked for one, and an interviewer scrambled to find and hastily open an IDE on the fly. • I was given a piece of partial code and asked to implement a class structure involving an abstract base class and a managing pipeline class. • I asked for clarification on the goal of the task and what I was expected to do. The response was simply that I needed to “figure it out.” • Clarifying questions received vague and misleading hints, which led me down an incorrect implementation path. • In hindsight, it was a very simple question. But during the interview — under pressure, without a clear explanation, and with misleading guidance — I couldn’t figure out what I was actually being asked to do. • As a result, I was unable to answer the first question, and the interview was abruptly ended after ~30 minutes with no feedback or opportunity to recover. Conclusion: Despite being labeled a Data Scientist role, the interview process focused almost entirely on OOD and class design — with minimal relevance to modeling, experimentation, or applied data work. Poor preparation, unclear expectations, and misleading communication made this a frustrating and unproductive experience. Recommendation: Clarify upfront whether the role is truly modeling-focused or leans heavily toward software engineering or ML infrastructure before investing your time.