The bank contains 47 new assignments: 20 Python, 4 SQL, 7 API/diagnostic, 6 projects, 5 prediction and 5 repair exercises. Two more relationship and aggregation exercises are in the first SQL workbook. All examples and data are synthetic.
Choose a format
| Format | When to use it |
|---|---|
| What happens? | Predict output and explain behavior before running code |
| Find and fix | Reproduce a defect, make a minimal patch and test it |
| Python | After language theory: implement, run and explain behavior |
| SQL | After JOIN and GROUP BY: schemas, boundaries and query plans |
| APIs and diagnostics | When you can build an API and want to test failure handling |
| Projects | For independent implementation and defending your decisions |
Where to start before an interview
Our archive review covers 335 interviews. The table shows missed answers and assessed answers per topic. A miss includes an incorrect, incomplete or absent answer. Ratings were made by AI. Methodology and limitations.
Scroll the table horizontally →
| Topic | Misses / assessed answers | Practice |
|---|---|---|
| Types and data structures | 187 / 459 | Open exercise |
| Concurrency and the GIL | 138 / 269 | Open exercise |
| Indexes | 117 / 237 | Open exercise |
| OOP | 87 / 161 | Open exercise |
| Iterators and generators | 78 / 155 | Open exercise |
| Message brokers | 87 / 149 | Open exercise |
How to practice
- Choose one assignment for your stage.
- Write down assumptions and expected outcomes.
- Implement a solution and edge-case checks.
- Compare with the criteria. Open hints after attempting it.
- Explain trade-offs aloud in two minutes, then repeat with different data.
Bring code, run commands, tests and a short design explanation to review. For diagnostic work, add measurements before and after the change.