Race setup
Choose a dataset, challenge style, difficulty, and time limit before starting the race.
Score
0
Speed and accuracy
Cleanliness
0%
Target: 90%
Issues remaining
0
Across all categories
Correct streak
0
Bonus after three
Mistakes
0
Sudden death limit: 3
Hints left
3
Hints cost points
Race clock
02:00
Cleaning progress
0 of 0 issues fixed
Rival progress
0%
Race track
Your cleaner advances as the dataset quality improves.
Blue: you
Purple: rival
Red: data errors
Cleaning tools
Select rows or cells, then apply a focused cleaning action.
Editable dataset
Click any unlocked cell to correct it manually.
0 records shown
Page 1 of 1
Keyboard: Ctrl+Z undo, Ctrl+Y redo, Ctrl+Enter validate.
Quality dashboard
Uniqueness0%
Validity0%
Consistency0%
Completeness0%
Label quality0%
Coach feedback
Start a race to receive validation guidance and learning feedback.
Issue summary
Performance history
Best scores and recent rounds are stored locally in this browser.
| Date | Theme | Mode | Difficulty | Score | Cleanliness | Time |
|---|
Learning summary
What the race teaches
Duplicates distort counts, invalid values break analysis, inconsistent formats block grouping, missing fields reduce completeness, and label errors weaken machine-learning models.
Recommended workflow
Profile the data, identify rules, fix high-impact errors, review duplicates carefully, validate changes, compare quality metrics, and preserve an audit trail.