Help Center

Each challenge provides a database with the tables and sample data you see in the schema panel. Write your query, hit Run, and the grader checks your output against the expected result set.

DataDriven uses SQLite syntax, which differs slightly from PostgreSQL or MySQL. Common gotchas: no ILIKE (use LIKE with LOWER()), no FULL OUTER JOIN (use UNION of LEFT and RIGHT), and DATE functions follow SQLite conventions. Check the schema panel for exact column types.

Yes. SQLite supports ROW_NUMBER(), RANK(), DENSE_RANK(), NTILE(), LAG(), LEAD(), and aggregate window functions with OVER(). QUALIFY is not supported, but you can wrap in a CTE and filter.

The grader checks your output against the expected result set. Common issues: column ordering matters, column aliases must match, and NULL sort order can differ. If you believe the grader is wrong, use the contact form and include your query.

Each execution gets a fresh, isolated environment. Write your solution, hit Run, and you will see stdout output and grading results. Challenges are designed to be solved with built-in Python.

The coding environment includes the Python standard library. Third-party libraries like pandas, numpy, or requests are not available. Challenges are designed to be solved with built-in Python only, which mirrors what most interviews expect.

Check for infinite loops, unnecessary nested iterations, or large data generation. If your solution is O(n^2) or worse on a large input, try optimizing your approach.

Challenge attempts, lesson progress, skill assessments, and streaks are stored securely in your account. Progress syncs across web and mobile. View your stats anytime on your profile page.

Make sure you are signing in with the same Google account you originally used. If you previously used email/password and then switched to Google (or vice versa), you may have created a second account. Contact us and we can merge them.

Yes. Challenge completions, lesson progress, and skill scores sync in real time as long as you are signed in with the same account on both platforms.

You get a timed coding round that mirrors a real interview. A prompt is revealed, you write your solution under time pressure, then an AI interviewer walks through follow-up questions, edge cases, and optimization discussions.

SQL (JOINs, window functions, CTEs, aggregations), Python (data structures, ETL patterns, file I/O), data modeling (star schemas, SCDs, normalization), and pipeline architecture (batch vs streaming, orchestration, idempotency).

Yes. You can re-attempt any interview challenge as many times as you want. Each attempt is tracked separately so you can see your improvement over time.

A new challenge is posted every day at midnight UTC. It rotates across SQL, Python, data modeling, and architecture topics. Completing daily challenges builds your streak and helps reinforce skills over time.

Yes. Every challenge is reverse-engineered from real interview loops at companies like Meta, Google, Amazon, Netflix, Stripe, and others. Topic weightings reflect what those companies actually test.

Challenges range from Foundation (entry-level) through Advanced (senior IC). Your onboarding placement test calibrates your starting point, and the adaptive system adjusts difficulty as you progress.

Yes. DataDriven is available on the App Store for iOS. The mobile app includes interactive drag-and-drop builders, touch-optimized challenges, and offline lesson reading.

Yes. Every feature is unlocked on mobile, same as the web app. No in-app purchases, no ads.

Not yet. The web app at datadriven.io works great on Android browsers in the meantime. We recommend Chrome for the best experience.

Most preferences are editable directly in Settings. For any other request relating to your stored information, you can reach out to support@datadriven.io and we will help where applicable.