What problem does it solve? Preparing for senior data-engineering technical screens requires realistic practice on Spark and Iceberg internals, but solo study lacks the pressure, probing follow-ups, and calibrated feedback of a real interview. ## Core Features & Use Cases - Live role-play mock interview: An interviewer persona (Staff DE or Senior IC) asks 8-12 questions mixing production scenarios (OOM at 3am, small-file explosions, traffic spikes) with direct internals questions (AQE, hidden partitioning, compaction, WAP). - Three drill modes: Spark-only, Iceberg-only, or Mixed, drawing from dedicated question banks in references/spark-questions.md and references/iceberg-questions.md. - Strict per-question feedback: Each answer is scored on correctness, depth, production realism, and trade-off awareness, with anti-patterns like tool_reflex and confident_wrong flagged. - Use Case: A senior DE candidate says "drill me on Spark and Iceberg" before a Netflix onsite; the skill runs a 45-60 minute mock, then delivers a final debrief with per-axis scores, top fixes, and knowledge-base gap updates. ## Quick Start Ask the assistant to start a mixed Spark and Iceberg mock interview using this skill and answer each question as it is posed.
