Data Engineer Coding Test — Scaffold

Working repo for a 1–2 hour data engineering coding test (SQL + Java + Python) for an academic-institution role. This is a ready-to-go scaffold, not a solved test — drop the real assignment's requirements into each track below and start solving.

Time-box plan (suggested)

• 5 min — read the prompt fully, note deliverables and constraints
• 45–50 min — SQL track (schema/queries)
• 30–35 min — Python track (ETL/analysis)
• 20–25 min — Java track (if required)
• 10 min — write up assumptions, run everything once more, commit

Layout

sql/       schema.sql, seed.sql, queries.sql, docker-compose.yml (local Postgres)
python/    src/ + tests/, pytest-ready
java/      Maven project, src/main + src/test, JUnit 5
docs/      scratch notes / assumptions while solving

SQL track

cd sql
docker compose up -d          # starts Postgres on localhost:5432 (db: testdb, user/pass: test)
psql postgresql://test:test@localhost:5432/testdb -f schema.sql
psql postgresql://test:test@localhost:5432/testdb -f seed.sql
psql postgresql://test:test@localhost:5432/testdb -f queries.sql

schema.sql ships a small academic domain (students, courses, enrollments) as a starting point — replace with whatever schema the test specifies.

Python track

cd python
python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
pytest -q

Java track

cd java
mvn -q test

VS Code

Open data-engineer-coding-test.code-workspace (multi-root: root, sql, python, java). It recommends the Python, Java Extension Pack, SQLTools (+Postgres driver) and Docker extensions, and points the Python interpreter at python/.venv.

Notes

Use docs/NOTES.md to jot assumptions, edge cases, and anything you'd ask a reviewer if you had time — worth including in the submission.