Launching today

ExSift

Launching today

Clean and transform data — no more manual cleanup or scripts

1 follower

Clean and transform data — no more manual cleanup or scripts

1 follower

Visit website

Stop wasting hours cleaning vendor data. Every company gets data from vendors, suppliers, and partners—each in a different format. ExSift solves this with Scenarios: define your output structure once, then upload any file. Its AI maps fields, normalizes formats, deduplicates, and builds relationships automatically. Reusable for future files, it handles catalogs, ETL pipelines, or partner data at scale. Result: clean, structured JSON or DB-ready data—no manual work, no scripts.

Free

Launch tags:Spreadsheets•Artificial Intelligence•E-Commerce

Launch Team

Anima - OnBrand Vibe Coding — Design-aware AI for modern product teams.

Design-aware AI for modern product teams.

Promoted

Maker

📌

Hey Product Hunt! 👋 I'm the maker of ExSift, and I'm excited to share it with you today. This started from a real frustration. We were ingesting product catalogs from 10+ suppliers — and every single one had a different format. Some used "Name", others "Product Title", others "Full Name". Prices came in as "12,99 USD", "$12.99", and "12.99EUR". We were spending days each month just cleaning data before we could use it. We built ExSift around a simple idea: you should only have to describe your data once. Define a Scenario — your expected fields and structure — and ExSift will map any incoming file to it, no matter how it arrives. No scripts. No manual cleanup. Ever again. A couple of things I'm especially proud of that we haven't seen elsewhere: Semantic classification. Upload your own taxonomy — your category tree, your tag structure, whatever you use — and ExSift automatically infers the right classification for each record based on its name, description, or any other field. No manual tagging, no lookup tables to maintain. VLookup. Beyond just building relationships between sheets, ExSift can enrich incoming records by matching them against reference datasets — think auto-resolving supplier codes to internal IDs, or pulling in pricing tiers from a master list. Under the hood it's a hybrid: AI for understanding and classifying, deterministic rules for consistent and reliable output. Flexibility without unpredictability. I'd love to hear from you — especially if you're dealing with messy vendor data, fragile ETL scripts, manual category tagging, or supplier onboarding headaches. Happy to answer any questions below! 🙏

Report

1d ago

Forum Threads

p/exsift

•

22h ago

Why is "Standard CSV" never actually standard? 🙃

I ve spent the last few weeks staring at vendor files, and I m convinced that "CSV" stands for Constantly Slightly Variable.

I was trying to ingest a few product catalogs recently. One supplier sends prices as 12.99 USD, the next uses $12,99, and the third just puts 1299 and expects you to guess the decimals. I ended up spending more time writing "one-off" cleanup scripts than actually using the data.

View all