Notes On Entity Resolution, MDM And Data For AI

Entity Resolution Software: 50+ Tools Compared

Over 50 entity resolution tools compared by entity types, match logic, blocking, and where they run: engines, MDM suites, CDPs, identity data and graphs.

November 10, 2026

A Human Corrected the AI. Then the Data Changed.

Master Data Management

A Human Corrected the AI. Then the Data Changed.

What happens when an AI agent resumes work after a human approves a decision? Lessons on data stewardship, agents and MDM from building Zingg 1.0 Enterprise.

November 3, 2026

Discovered by Agents: What Zingg's Telemetry Shows About Agentic Entity Resolution

An AI agent ran Zingg's match phase every few minutes, unprompted. What Zingg's telemetry shows and what makes open source entity resolution agent-friendly.

October 30, 2026

Top 10 Open Source Entity Resolution Tools (2026)

From active-learning tools to probabilistic linkers, see which open source entity resolution tool actually fits your data problem in 2026

August 25, 2026

Your data doesn’t live in the “right” format. Neither should your entity resolution

Your data doesn’t live in the “right” format. Neither should your entity resolution

Zingg reads and writes Excel files natively via Spark, no CSV or Parquet conversion needed. See how entity resolution works directly on .xlsx data

August 11, 2026

Zingg 0.7.0: A More Stable, More Usable Zingg

Zingg 0.7.0: A More Stable, More Usable Zingg

Zingg 0.7.0 improves error handling for faster debugging, restructures Databricks/Fabric notebooks into modular steps, and upgrades core dependencies.

August 10, 2026

Zingg Python API For Entity Resolution Just Hit the Top 10% of PyPI

Zingg Python API For Entity Resolution Just Hit the Top 10% of PyPI

Zingg's open-source entity resolution engine now ranks in PyPI's top 10% with 77K+ downloads, 591 PRs, and 573 issues. Here's what that milestone means.

August 4, 2026

The Fellegi-Sunter Model: Five Structural Limits Modern Data Can No Longer Ignore

The Fellegi-Sunter Model: Five Structural Limits Modern Data Can No Longer Ignore

Fellegi-Sunter was designed for 1969 data. Explore 5 critical limitations modern data teams can't ignore and how ML-based entity resolution fixes each one.

June 1, 2026

The Limits of Entity Centric Matching and a better way forward

The Limits of Entity Centric Matching and a better way forward

Entity centric matching sounds smart but falls short at scale. Explore 5 key limitations around auditability, accuracy, and scalability—and what works better.

June 1, 2026

Evolution of MDM

From rule-based hubs to AI-powered entity resolution, see how MDM tools evolved to handle modern data complexity and deliver trusted golden records at scale.

May 18, 2026

Zingg 0.6.0: Databricks Unity Catalog Support, Spark 3.5.5, and a Lighter Python Wheel

Zingg 0.6.0: Databricks Unity Catalog Support, Spark 3.5.5, and a Lighter Python Wheel

Zingg 0.6.0 releases UCPipe for Databricks Unity Catalog and Delta table support, Spark 3.5.5 upgrade, better error handling, and a lighter Python wheel.

May 8, 2026

Identity Resolution API: How to Resolve Entities Programmatically with Zingg

Identity Resolution API: How to Resolve Entities Programmatically with Zingg

Use Zingg's Python identity resolution API to define fields, label training data, train your model, and run entity matching on Databricks and Microsoft Fabric

April 23, 2026

Build vs. Buy Entity Resolution: Why "Just Re-Run It" Will Break Your Stack

Thinking of building entity resolution in-house? Learn why incremental matching, stable IDs, and human feedback make it harder than most data teams expect.

April 21, 2026

Build vs Buy: Identity Resolution on BigQuery vs a Packaged CDP

Build vs Buy: Identity Resolution on BigQuery vs a Packaged CDP

Packaged CDPs take 12–24 months and cost millions. See why GCP-native teams choose identity resolution on BigQuery with Zingg over a packaged CDP solution.

Build vs Buy: Identity Resolution on AWS vs a Packaged CDP

Build vs Buy: Identity Resolution on AWS vs a Packaged CDP

Packaged CDPs promise identity resolution on infrastructure you don't control. See why AWS teams resolve identities with Zingg inside their own data platform.

Build vs Buy: Identity Resolution on Microsoft Fabric vs a Packaged CDP

Build vs Buy: Identity Resolution on Microsoft Fabric vs a Packaged CDP

Fabric teams don't need a separate CDP for identity resolution. Zingg runs inside Fabric and writes a persistent ZINGG_ID back to your Lakehouse.

Build vs Buy: Identity Resolution on Snowflake vs a Packaged CDP

Build vs Buy: Identity Resolution on Snowflake vs a Packaged CDP

Packaged CDPs take months to implement and cost millions. Zingg resolves identities natively in Snowflake and writes a persistent ZINGG_ID to your tables.

Build vs Buy: Identity Resolution on Databricks vs a Packaged CDP

Build vs Buy: Identity Resolution on Databricks vs a Packaged CDP

The real pros and cons of building identity resolution natively on Databricks versus buying a packaged CDP, and what it means for your martech stack.

Entity Resolution at Scale, Part 5: Incremental Flow and Living Clusters

Batch matching is only half the problem. Learn how incremental entity resolution keeps your identity graph current as records arrive, change, and split.

Entity Resolution at Scale, Part 4: Thresholds, Scores, and Active Learning

Setting match thresholds by hand is a losing game. See how probabilistic scores and active learning find the right operating point from labeled pairs.

Entity Resolution at Scale, Part 3: Blocking: Making Billion-Record Matching Tractable

Comparing every record pair is computationally impossible at scale. Learn how blocking strategies reduce comparisons by 99%+ without missing true matches.

Entity Resolution at Scale, Part 2: Stopwords, Normalization, and Domain Tuning

Noise in your data kills matching accuracy before it starts. See how stopword removal, normalization and domain tuning fix it at the source.

Entity Resolution at Scale, Part 1: Why One Similarity Metric Is Never Enough

Single similarity metrics fail on real data. Why production entity resolution needs multiple metrics, field-level weighting and composite scoring.

We built a small bot to track what's happening with Zingg every day

We built a small bot to track what's happening with Zingg every day

Automation scripts to understand githu and google analytics for Zingg open surce entitu y resolution

April 14, 2026

Master Data Management

The Modern Guide to Master Data Management with AI and Why Legacy MDM Is Being Replaced

Legacy MDM was built before cloud data platforms. What master data management looks like now, where AI fits, and why data teams are replacing legacy hubs.

April 13, 2026

Building an Entity-Resolved Knowledge Graph: Zingg and Neo4j Together

Building an Entity-Resolved Knowledge Graph: Zingg and Neo4j Together

Entity resolution cleans the nodes; Neo4j maps the relationships. How to connect Zingg and Neo4j so each entity appears once in your knowledge graph.

April 10, 2026

Beyond the CDP: Building a Real Customer 360 on Your Data Platform

Beyond the CDP: Building a Real Customer 360 on Your Data Platform

Packaged CDPs promise Customer 360 but deliver integration projects. Build a persistent customer view in Databricks, Snowflake, Fabric or BigQuery instead.

How We Measure Open Source Adoption at Zingg: A Frugal Approach

How We Measure Open Source Adoption at Zingg: A Frugal Approach

GitHub stars are a noisy signal. How Zingg measures what matters for an open source project: usage, retention and the teams solving real problems.

Deterministic vs. Probabilistic Matching: Why Enterprise Entity Resolution Needs Both

Deterministic vs. Probabilistic Matching: Why Enterprise Entity Resolution Needs Both

Deterministic and probabilistic matching solve different problems. Why production entity resolution needs both, and how Zingg handles each.

Incremental Identity Resolution: Keeping Your Entity Graph Current as Data Changes

A one-time entity resolution run is fine for a report, not a business. Keep your entity graph accurate as records arrive, change and split.

The ZINGG_ID: A Persistent, Load-Bearing Identifier for Your Entity Graph

The ZINGG_ID: A Persistent, Load-Bearing Identifier for Your Entity Graph

Clusters say which records belong together today. ZINGG_ID says how to refer to that entity tomorrow. How persistent identifiers work in production.

April 8, 2026

Snowpark Performance Tuning for Entity Resolution: From 12 Hours to 30 Minutes

Snowpark Performance Tuning for Entity Resolution: From 12 Hours to 30 Minutes

Our baseline Snowflake entity resolution run took 12 hours on 500K records. The bottlenecks we found and how Snowpark tuning cut it to 30 minutes.

April 8, 2026

How We Built Native Snowflake Entity Resolution with Snowpark

How We Built Native Snowflake Entity Resolution with Snowpark

Zingg used to run on Spark next to Snowflake. Now it runs inside Snowflake via Snowpark. How we built it and what it means for warehouse-only teams.

April 8, 2026

Record Linkage vs. Entity Resolution vs. Data Deduplication: What's the Difference?

Record linkage, entity resolution and deduplication are related but distinct. What each does, when to use it, and how they fit on your data platform.

March 26, 2026

Why Entity Resolution Is Harder Than It Looks

String matching, similarity metrics, thresholds, scale and clustering: five problems that make entity resolution look simple and stall in-house builds.

March 24, 2026

When Customers Grow, We Grow: Redica’s Journey with Zingg

When Customers Grow, We Grow: Redica’s Journey with Zingg

Redica Systems was an early adopter of Zingg's open source entity resolution. Congratulations on their fundraise, and how we grew together.

May 17, 2025

We Were Just Solving a Problem. Then We Saw 1,000 Stars.

We Were Just Solving a Problem. Then We Saw 1,000 Stars.

The Zingg open source entity resolution repository crossed 1,000 GitHub stars. The story behind the number and what we learned from our users.

May 8, 2025

Master Data Management

Master Data Management Eats AI for Breakfast, or Does It?

AI fails without clean master data, but that doesn't mean legacy MDM wins. How MDM and AI relate, and why the answer is neither legacy MDM nor nothing.

May 8, 2025

Deduplicating Records, Is Machine Learning the Answer?

Deduplicating Records, Is Machine Learning the Answer?

Rule-based deduplication breaks on real data variation. When machine learning beats hand-written rules for deduplication, and what a practical approach is.

May 8, 2025

Lessons Learnt While Integrating Zingg's Open Source Entity Resolution Python APIs With Microsoft Fabric

Lessons Learnt While Integrating Zingg's Open Source Entity Resolution Python APIs With Microsoft Fabric

Running Zingg's Python entity resolution API on Microsoft Fabric: the Spark config, notebook and Delta table gotchas we hit, and how we fixed them.

April 12, 2025

Kickstarting Zingg Community Events!

Zingg community events start with Sem Sinchenko, Zingg contributor and Apache GraphAr PPMC member, on how Spark Connect changes distributed computing.

September 19, 2024

Zingg Joins The AWS SaaS Spotlight Program!

Zingg Joins The AWS SaaS Spotlight Program!

Zingg was selected for the AWS SaaS Spotlight Program for early-stage startups building on AWS. What it means and what we're working on.

September 17, 2024

Welcoming Shelomo Dobkin to Zingg as an Advisor

Welcoming Shelomo Dobkin to Zingg as an Advisor

Shelomo Dobkin, Director of Product Development at Orthodox Union, joins Zingg as an advisor, bringing experience in data management at scale.

July 31, 2024

Identity Resolution

1000 Downloads of Zingg Identity Resolution Python!

The Zingg identity resolution Python package passed 1,000 downloads. Why we built Zingg and how teams use it for customers, patients and more.

September 22, 2023

Deep Dive into FHIR and Duplicate Data Challenges in Healthcare Data and a Solution Using Zingg

Deep Dive into FHIR and Duplicate Data Challenges in Healthcare Data and a Solution Using Zingg

FHIR promises interoperability but doesn't fix duplicate patients. How entity resolution matches patient records across systems into one trusted view.

June 8, 2023

The Comprehensive Resource on Composable Customer Data Platforms and Identity Resolution

A curated guide to composable customer data platforms, the role of identity resolution, and why Snowflake and Databricks teams replace packaged CDPs.

May 22, 2023

Zingg Community Meetup on 2nd September 2022

Identity Resolution

Zingg Community Meetup on 2nd September 2022

Zingg users shared how they use open source entity resolution for Customer 360 and spend analytics at our September 2022 community meetup.

September 5, 2022

My internship experience with Zingg

Raviraj Baraiya, an IIT Kharagpur senior, on interning at Zingg: working on open source entity resolution and what a startup internship teaches you.

July 22, 2022

Zingg en France

Zingg joins Bpifrance's Data Summer Days 2022 in Paris to show how large-scale entity resolution supports AML, KYC and fraud work in financial services.

July 9, 2022