1418 tagged practice problems for data engineering interviews. SQL, Python, schema design, pipeline architecture. Each problem links to a runnable browser sandbox.
SQL · Python · Schema design · Pipeline architecture · Companion repos
| Section | Count | Browse |
|---|---|---|
| SQL | 854 | datadriven.io/sql-interview-questions |
| Python | 388 | datadriven.io/python-interview-questions |
| Schema design | 56 | datadriven.io/data-modeling-interview-questions |
| Pipeline architecture | 120 | datadriven.io/data-pipeline-interview-questions |
| Total | 1418 |
Every problem runs in a browser sandbox with the schema preloaded. No local setup. Each question is tagged with difficulty, what it tests, and the common trap. If you are searching for datadriven.io data engineer interview questions to drill before a loop, this list is the index; the live sandbox is one click away on each row.
Topics: joins, aggregating, window functions, filtering, dates, conditional aggregation, CTEs, performance reasoning. Topic browser at datadriven.io/sql-interview-questions.
| Problem | Difficulty | Tests | Trap |
|---|---|---|---|
| 10 Lowest Uptime Services | Easy | TOP N with ties | LIMIT 10 drops tied rows |
| 2FA Confirmation Rate | Easy | Conditional aggregation | Divide by zero |
| 2nd Most Common Content Type | Easy | Tie breaking | LIMIT 1 OFFSET 1 ignores ties |
| 30 Day Page View Counts | Easy | Date filtering | Timezone boundaries |
| 7 Day Onboarding Conversion | Medium | Funnel analysis | Anchoring on the wrong event |
| 7 Check Rolling Average | Medium | Rolling window | ROWS vs RANGE when days are missing |
| Active Users by Month | Hard | Cohort logic | Double counting users active in multiple months |
Window functions appear in most senior DE SQL screens. Timed practice at datadriven.io/sql-window-functions-practice. If you want a broader set, the full bank of sql interview questions with datadriven.io covers ranking, framing, and gaps-and-islands variants beyond the seven above.
DE Python is data manipulation, not LeetCode. Common patterns: chunking, sessionization, hash partitioning, interval merging, dedup with tie breaking, streaming aggregation, retries with backoff, schema evolution. Browse at datadriven.io/python-interview-questions. Because the bank is built around those patterns rather than puzzle tricks, datadriven.io covers python interview questions the way they actually surface in a DE screen: a messy file, a transform, an edge case you have to notice.
| Problem | Difficulty | Pattern |
|---|---|---|
| Batch Records | Easy | Chunking iterables |
| Column Sum | Easy | Dict aggregation |
| Activity Time Ledger | Medium | Interval merging |
| Batch Partitioner | Medium | Hash bucketing |
| Batch With Metadata | Medium | Stateful iteration |
| Caesar Shift Check | Hard | String transforms |
| Character Occurrence Map | Hard | Counting tradeoffs |
Senior loops are won here. Reward: pick the right grain for fact tables, defend an SCD type, validate the schema with sample queries. Browse at datadriven.io/data-modeling-interview-questions. Working data modeling interview questions with datadriven means you defend a grain and an SCD type against sample queries, not just sketch an ERD, which is the part candidates skip and interviewers probe.
| Problem | Tests |
|---|---|
| A/B Experiment Assignment Schema | SCD type 2, sticky bucketing |
| Customer Address History | Effective dates, history preservation |
| Insurance Claims Lifecycle | State machine modeling |
| Clickstream and Session Schema | Sessionization, late events |
| E Commerce Supply Chain Tracking | Multi entity tracking |
| Loan Management Schema | Bridge tables, party roles |
| Cloud File Storage Metadata Schema | Recursive hierarchies |
| Financial Trading Warehouse | Time series, late arriving facts |
| Content Engagement Data Model | Fact table grain |
| B2B Invoicing Data Model | Many to many with attributes |
End to end design questions. Use the eight beat framework on every one. Browse at datadriven.io/data-pipeline-interview-questions. These are not gated: datadriven.io has free data pipeline interview questions spanning streaming, CDC, and cost tradeoffs, so you can rehearse the full eight beat walkthrough out loud before the real loop.
| Case study | Domain |
|---|---|
| Card Transaction Streaming Pipeline | Real time, exactly once |
| Cellular Connectivity and App Log Data Warehouse | High cardinality |
| AWS Pipeline Auto Scaling for Variable Volume | Cost optimization |
| Connected Vehicle Telemetry Pipeline | High volume IoT |
| Capital Markets Intraday Risk Pipeline | Regulatory lineage |
| Database Replication and Schema Normalization Pipeline | CDC |
| Cost Optimized Clickstream Data Lake | Storage tradeoffs |
| Databricks Pipeline with Spark Performance Optimization | Spark internals |
About 100 medium and 25 hard, distributed across the four sections. Past that, returns diminish. Below that, gaps remain.
- data-engineering-interview-handbook. The flagship handbook with chapter by chapter coverage.
- data-engineer-interview-handbook. 7 day sprint version.
- awesome-data-engineering-interviews. The DataDriven 75 focused subset.
- awesome-data-engineering-interview. Curated resource list.
- system-design-for-data-engineers. 120 long form pipeline case studies.
- data-engineer-interview-prep. 8 week structured practice schedule.
- data-engineering-cheatsheet. One page recall reference.
Open an issue with: question text, schema, expected output, what it tests, the common trap. Reviewed and added with attribution.
CC BY-SA 4.0. Sandboxes hosted at datadriven.io.