Deloitte Data Engineer Interview Loop
Deloitte's data engineer loop is compact and often happens in one block. A recruiter call covers the role and salary. Then a technical step, either a coding test or a live interview with a senior engineer on SQL, Python and Spark. Then a round with a manager that walks through the architecture of your current project, sometimes drawn on paper. The last conversation is with a senior manager or partner about why Deloitte and what kind of project you want.
What decides it
Being able to explain a pipeline you built, from source to dashboard, to someone who will ask why at every step. Candidates who know the tools but cannot draw the system tend to stop at the manager round.
Practice by round
SQL and Python on timesheet data: idle employees in a month, hours logged after project end, duplicate timesheet rows, orphan timesheet entries, headcount by practice and level. These are the data quality checks a consultant runs on a client's export, and the loop expects you to write them without hesitation.
Harder SQL and Python: top clients by revenue, time differences between events, and Python puzzles like reversing letters while keeping word order or building a deque from a linked list. Spark questions are about reading a slow job and saying what you would change.
| Question | Difficulty |
|---|---|
| Difference Between Times | Medium |
| Time from 10th Runner | Medium |
| Population Density | Medium |
| Reverse Letters, Preserve Words | Medium |
| Variable Swap Challenge | Medium |
| Top Clients by Revenue in 2025 | Medium |
| Prime Numbers Below N | Hard |
| Linked List Deque Creation | Hard |
Design and the project. The case on a cost-effective cloud batch pipeline is the conversation the architecture round turns into. Layered storage, incremental runs, idempotent writes, what to do with late data. Database normalization and table joins cover the fundamentals, and the data project takes a problem end to end.
| Case study | Difficulty |
|---|---|
| Database Normalization | Medium |
| Scalable Data Science | Medium |
| Table Joins | Medium |
| Cost-Effective Cloud Batch Pipeline | Medium |
| Data project | Difficulty |
|---|---|
| Financial Health Check | Hard |