Your cohort ships real pipelines before graduation.
Hands-on data engineering labs on a live lakehouse. Students code in the browser, every attempt is autograded check by check, and you watch the whole class progress in real time.
Teaching data engineering shouldn’t mean running an IT department.
Drivers, Docker, cloud credits, “it works on my machine”.
✓Minute one, everyone codes. Browser only: nothing to install, no cloud bill.
40 students × 12 labs × 2 attempts is 96 hours of review.
✓Each submission graded in seconds against hidden test cases, with partial credit.
Toy SELECTs on clean CSVs don’t survive a first on-call.
✓Labs are real incidents: late data, duplicates, idempotent re-runs, SCD type 2.
What your next lab session feels like.
Same room, same students. You just stop being the bottleneck.
- 00:00Open the assignment
The window opens and all 40 students start. Nobody raises a hand about setup.
- 00:05Students explore real tables
Iceberg tables on Trino, a table explorer and scratch queries on the sample data.
- 00:20You see who’s stuck
The learners × labs grid updates live. Walk straight to the amber cells.
- 00:45Feedback without waiting for you
Every Submit runs hidden cases. Students see which check failed and why, never the answer.
- 01:30The window closes, grades freeze
Best attempt counts. Every attempt is kept with its code for review.
Everyone in the room gets something.
Outcomes you can show. A cost per student you can defend.
- Graduates who have built incremental loads, deduplication and SCD type 2 on Iceberg and Trino, the stack hiring managers ask about.
- One view per cohort: who started, who completed every lab, the average score. Ready for your next program review.
- Build a track once, run it for every intake and campus. Dates live on the assignment, not the content.
- $3 per student per month. No cloud accounts to manage, no infrastructure team, no surprise bills.
Built for how programs actually run.
Everything marked Available works today. The rest is on the roadmap, and pilot schools shape it.
Trino SQL and PySpark on Apache Iceberg tables. The tools of the job, not a simulator.
Hidden test cases and weighted checks. Students learn from which check failed, not just a red X.
Group your students, order labs into modules, give each cohort its own time window.
A learners × labs grid and a submissions feed with every attempt’s code and results.
Each submission runs isolated with a short-lived identity. Nobody sees anyone else’s work or data.
Graded on cases students never see. Every attempt is timestamped and kept with its code.
Runs queue fairly per student. An infrastructure hiccup before your code starts is retried automatically.
Student data stays in the EU. A data processing agreement is available on request.
Students use the school account they already have. Campus SSO for larger programs.
Launch labs from Canvas, Moodle or Blackboard and sync grades back.
Turn your datasets into graded scenarios, visible only to your institution.
A shareable record of what each graduate built, for their CV and your placement stats.
An AI teaching assistant that sits in your live session.
Not a chatbot that does the homework: a co-teacher that reads the whole room and keeps students thinking.
Clusters the class’s failing checks live, with the likely misconception. Fix it for 12 students at once.
Socratic hints that never reveal the solution. You decide how far they go.
Describe a learning objective; get a draft scenario, datasets and checks to review and publish.
After class: who is at risk, and which concept to revisit next time.
Your second run inserts rows the first one already wrote. What makes a row the same order, and how could MERGE use it?
You’re probably thinking…
What your next cohort costs, and what it gives back.
Assumes 6 min to review one submission by hand and 2 attempts per lab.
Priced per student. Teachers never pay.
Start with one class free for a term. Pay only when it works for you; cancel any term.
- ✓One class, up to 30 students
- ✓Every Classroom feature
- ✓Teachers and TAs free
- ✓Onboarding call and your first custom lab
- ✓Unlimited cohorts, tracks and assignments
- ✓All published labs: Trino SQL and PySpark
- ✓Live progress grid and submission review
- ✓EU hosting, GDPR data processing agreement
- ✓Email support within one business day
- ✓Everything in Classroom
- ✓Campus SSO and LMS integration as they ship
- ✓Private labs built on your datasets
- ✓Dedicated EU instance on request
- ✓Early access to the AI teaching assistant
Prices in USD, excluding VAT. Teachers, TAs and staff accounts are always free.
Run your next lab on Shuffleet.
Bring one class. We set up the cohort with you, your students start the same day, and the first term is on us.