Upload a messy CSV/Excel file and it:
Normalizes headers (snake_case)
Fixes inconsistent data types
Removes duplicates
Strips hidden Unicode junk
Standardizes formatting
The pipeline is deterministic and idempotent. Files are stored ephemerally and auto-purged.
The interesting part: I used Google Gemini heavily as an engineering collaborator — not just for boilerplate, but for designing parts of the cleaning logic and app structure.
Would love feedback on:
Edge cases I might be missing
Performance concerns at scale
Better approaches to deterministic cleaning
Live demo: https://auto-table.com
Write-up: https://dev.to/voxdroid/autotable-my-one-click-spreadsheet-c...