Went through the full Target data engineering interview loop in the spring. Posting this because when I searched for it I found almost nothing recent. Here's the full breakdown.
Online Assessment. Standard HackerRank format. Two SQL problems (moderate complexity: window functions, CTEs, a self-join situation) and one Python/pandas problem around cleaning a malformed dataset. Total 75 minutes. The SQL was genuinely used-in-production level, not textbook. One question had a "find the second most recent order per customer" structure that trips people up if they're not comfortable with ROW_NUMBER() and partitioning.
Hiring Manager Screen (30 min). Mostly background and fit. They asked about my experience with batch vs streaming pipelines, which cloud platforms I'd worked on (they use Azure heavily at Target, worth knowing), and how I approach data quality issues in production pipelines.
Technical Deep-Dive Round (60 min). Two parts. First half was design: walk me through how you'd build an end-to-end pipeline for ingesting daily POS (point of sale) transaction data from 2,000 stores, transforming it, and making it available for analytics. We got into partitioning strategy, handling late-arriving data, idempotency, schema evolution. Real stuff.
Second half was live SQL: given a schema they provided (orders, products, stores, customers), write queries for 3 problems. One aggregation with filtering, one ranking with window functions, one that required a multi-step CTE. Time-pressured but reasonable.
Behavioral Round. Standard STAR questions focused on how you've handled data quality incidents, worked with stakeholders to define data requirements, and dealt with pipelines breaking in production.
Total process was about 5 weeks from OA to offer. My offer was in the mid-to-high range for a senior data engineer in Minneapolis. Not Bay Area numbers but very livable for the market.