🇮🇳 Independence Day Special Offer – Learn More, Pay Less!
ETL Pipeline Simulator
Run a real ETL job, one stage at a time. Every stage executes actual SQL against the live database — extract, stage, validate, transform, load, reconcile — and you can pause, step back and replay.
- 01
Nothing here is animated theatre. Each stage runs the SQL shown next to it, and the row counts come back from the database.
- 02
src_sales arrives dirty. Watch the reject count at the validation stage and the row count difference at load.
- 03
Step backwards and the job re-runs deterministically from the top — the only honest way to rewind a data pipeline.
Job control
0 of 6 stages complete
Extracted
0
rows in src_sales
Staged
0
rows landed in stg_sales
Rejected
0
failed a validation rule
Loaded
0
rows published to tgt_sales
Stage 1: Extract
The exact SQL this stage executes
SELECT COUNT(*) AS extracted_rows,
MIN(sale_date) AS first_date,
MAX(sale_date) AS last_date
FROM src_sales;Reject table
Appears once the validation stage has run
Run the job as far as stage 3 and every rejected row shows up here with its reason.
Run something and the rows land here.