Build
on the canvasDrag real services onto the canvas, wire the DAG, and configure every node — write mode, triggers, delivery guarantees.
data engineering · beginner → senior
Master the Data EngineeringSystems.
From your first query to full system design — practice the way senior data engineers actually work.
- 0
- flagship studios
- 0
- foundation modules
- 0
- decision loop
Flagship Studios
studio 01
Data Pipeline
Draw the DAG, configure every node, route the failures.
Open the pipeline studio studio 02Data Modelling
Set the grain, the keys and the history. Then defend it.
Open the modelling studioKnowledgebase Library
SQL
Reason about grain, joins, windows, and query cost.
Python
Write transformations that are readable, tested, and correct.
Spark
Understand shuffles, skew, memory, partitions, and job behavior.
Data Modelling
Choose grain, keys, facts, dimensions, history, and ownership.
Data Pipeline
Plan ingestion, orchestration, idempotency, and recovery.
the loop
Build. Review. Defend.
Every scenario runs the same three beats — the loop a senior engineer runs in a design review, and the one an interviewer walks you through.
Review
code decides0
13checks run10passed3to improve
structure 100requirements 91reliability 82
✓Bad events route to a dead-letter path instead of blocking the stream.
!A sink write isn't idempotent — a replay would double-count.
A deterministic engine scores structure, requirements and reliability, lists what's working, and ranks what to fix first. Then the AI narrates it — grounded in your graph.
Defend
in your wordsProbe · interview
3 question(s) left · graded on the concepts you coverA seller edits a listing's price the day after an order ships. When you query that order's total — whose price does the row show, and why?
Then the follow-ups, one probe at a time, on the model you actually drew — the same questions that decide a senior interview.
start
Your first system is one click away.
Pick a scenario, draw it, and find out what a reviewer would say.