benzsevern

benzsevern / goldenmatch

Public

Entity resolution toolkit -- deduplicate records, match across sources, create golden records. 97% F1 on structured data, LLM scoring for products. Polars-native, 7800 rec/s, zero-config CLI.

pypi.orgprojectgoldenmatch data-engineering data-quality deduplication entity-resolution fellegi-sunter
15
1
100% credibility
Found Mar 23, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

GoldenMatch is an open-source Python toolkit for deduplicating records, matching entities across datasets, and generating golden records using fuzzy matching, blocking strategies, and optional AI enhancements.

How It Works

1
🔍 Discover GoldenMatch

You hear about a simple tool that cleans up messy lists by finding and merging duplicates automatically.

2
📦 Get it ready

Install the tool with one easy command and run the friendly setup helper to connect any helpers like AI thinkers if you want.

3
📁 Pick your file

Choose your messy customer or product list, and it smartly guesses how to clean it up.

4
✨ See duplicates instantly

The golden screen lights up showing groups of matching records with easy sliders to tweak and review.

5
✅ Approve and tune

Quickly check borderline matches, adjust confidence levels, and confirm what to keep.

🎉 Get your clean list

Export perfect golden records with no duplicates, ready to use, saving you hours of manual work.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 15 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is goldenmatch?

GoldenMatch is a Python toolkit tackling the entity resolution problem—deduplicating messy records, matching across sources like CSVs or databases, and building clean golden records. Drop a file on the zero-config CLI (`goldenmatch dedupe customers.csv`), and it auto-detects columns, blocks candidates with Polars at 7800 rec/s, scores via fuzzy/embedding methods for 97% F1 on structured data, then launches a TUI for live threshold tweaks and golden record previews. Add LLM scoring for trickier product catalogs, or sync live Postgres tables incrementally.

Why is it gaining traction?

It crushes entity resolution github alternatives like dedupe or Splink with true zero-config on files/databases, no training labels needed, plus a gold-themed TUI, REST API for real-time matching, and domain packs for retail/healthcare. Benchmarks show 97% F1 on DBLP-ACM at laptop speeds, LLM boosting Abt-Buy products from 44% to 72% F1 for pennies—users love the interactive review queue and rollback without Spark/GPU hassles.

Who should use this?

Data engineers merging duplicate CRM exports from Salesforce/HubSpot into golden customer records, analysts cleaning entity resolution datasets before ML pipelines, or backend devs embedding an entity resolution API into apps for real-time deduping. Ideal for teams facing the entity resolution pipeline grind on 10K-10M records without wanting custom blocking graphs or weeks of labeling.

Verdict

Grab it for fast entity resolution github workflows—CLI/TUI shine for quick wins, with solid benchmarks and 855 passing tests signaling quality despite 15 stars and 1.0% credibility. Early beta means watch for rough edges on massive scales, but MIT license and Colab demo make trialing golden match a no-brainer.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.