---
title: "Atlan - The Context Layer for AI"
url: "https://atlan.com/"
description: "The missing context layer for enterprise AI. Atlan gives every AI agent the data graph, business logic, and governance to act on trusted data."
keywords: "AI agents, metadata management, data governance, context layer, enterprise data, AI context, data catalog, Atlan"
---

> Atlan is hosting Context Conference, bringing together the leaders and builders at the frontier of giving AI the context it needs to understand their business. It runs online on October 28, 2026, from 11:00 AM to 2:00 PM ET. Atlan co-founder Prukalpa Sankar opens and closes the day. Leaders from AstraZeneca, BNY and Verizon share why they invest in context and what they get from it. Registrants get early access to The AI Context Gap, a new study from MIT Technology Review Insights. Register: https://atlan.com/context-conference/

Atlan is the context layer for AI. **Your AI doesn't know your business. Let's fix that.** Build a shared understanding of your data, your business logic, and your institutional knowledge, and make it available to every AI tool you run.

- Talk to sales / book a demo: https://atlan.com/forms/talk-to-sales-contact/
- Watch the context layer demo: https://atlan.com/context-layer-demo/

## How the context layer fits

The hero diagram shows three tiers:

- **Interfaces and agents** (consume context)
  - Custom agents: Salesforce Agentforce, Snowflake, AI assistants
  - Vertical agents: Decagon, Sierra, Writer
  - General-purpose agents: Anthropic Claude, OpenAI
  - Tools: Slack, Microsoft Teams, Jira
- **Open and portable context: the Enterprise Context Layer**, packaged as Context Repos
  - **AI-Ready Data**: the integrated, trusted, and AI-ready representation of an enterprise's data and knowledge assets. See [Connectors](https://atlan.com/connectors/) and [Data Lineage](https://atlan.com/data-lineage/).
  - **Semantics and Ontology**: the map of the business: its entities, metrics, and relationships. See [Context Agents](https://atlan.com/context-agents/).
  - **Agent Skills**: reusable, versioned, testable units of procedural knowledge. See [Context Engineering Studio](https://atlan.com/context-engineering-studio/).
- **Business systems** (sources of context)
  - Systems of record: Salesforce, SAP, HubSpot
  - Systems of semantics: Looker, Microsoft Power BI, Tableau
  - Systems of data: Snowflake, Databricks, Google BigQuery, Google Cloud
  - Systems of knowledge: Confluence, Microsoft SharePoint, Google Drive

## Trusted by AI-forward enterprises

Customers shown: Mastercard, HubSpot, Zoom, Dropbox, Autodesk, Nasdaq, Fox, PPG, GitLab, Virgin Media O2, Unilever, Workday, Elastic, NHS, Affirm, General Motors, easyJet, Medtronic, New York Life, Grainger. All stories: https://atlan.com/customers/

Customer quotes on the logo wall:

- "AI initiatives require more context than ever. Atlan's metadata lakehouse is configurable, intuitive, and able to scale to hundreds of millions of assets." — Andrew Reiskind, Chief Data Officer, Mastercard. [Mastercard's story](https://atlan.com/regovern-watch-center/mastercard-context-by-design/)
- "Atlan's open and extensible foundation helps our technical team build applications to improve customer experiences in the AI era." — Ashfaq Mohiuddin, VP, Enterprise Data & AI, HubSpot
- "We have Atlan as the metadata plane across our tech stack, independent of where technology is. We have one place to define the data, understand what it means, and where it comes from" — Oliver Gomes, VP, Analytics & Strategy, Fox. [Fox's story](https://atlan.com/customers/fox-governance-atlan/)
- "Atlan has been a partner since day 1 in our journey in just one yer, we have onboarded 6,000 people. Conversational analytics will go even further to reach every employee." — Mauro Flores, Executive Vice President, Data Democratization, Virgin Media O2. [VMO2's story](https://atlan.com/regovern-watch-center/virgin-media-o2-context-for-all/)
- "Atlan captures Workday's shared language to be leveraged by AI via its MCP server. As part of Atlan's AI labs, we're co-building the semantic layer that AI needs." — Joe DosSantos, VP, Enterprise Data and Analytics, Workday. [Workday's story](https://atlan.com/regovern-watch-center/workday-context-as-culture/)
- "We're focused on how we can scale context development as much as possible, and where can we leverage Atlan AI to build the most robust definitions across our data estate and make sure we're propagating it upstream and downstream." — Takashi Ueki, Head of Enterprise Data & Analytics, Elastic. [Elastic's story](https://atlan.com/regovern-watch-center/hidden-strategies-behind-ai-ready-tech-pioneers/)
- "Atlan has been a really good partner in helping us figure out how to register AI models and applications, and what metadata to put in place to meet the transparency requirements that [AI Governance] legislation asks for." — Sherri Adame, Data Governance Lead, General Motors. [General Motors' story](https://atlan.com/regovern-watch-center/contracts-context-and-cars-general-motors/)

## Context Conference

The conference for teams teaching AI their business. October 28 / 11 AM ET / Virtual. Speakers shown: Jaya Gupta (Foundation Capital), Josh Klahr (Snowflake), Cindy Hoots (AstraZeneca), Bob Muglia (Snowflake), Leigh-Ann Russell (BNY), Lonne Jaffe (Insight Partners), Prajakta Damle (Google), Prukalpa Sankar (Atlan). Details and registration: https://atlan.com/context-conference/

## The Observation: enterprise AI fails not because of the model, but because of missing context

We've spent years studying how enterprises deploy AI agents. The pattern is consistent: teams build impressive prototypes, but hit a wall when moving to production.

The wall isn't the models. It's that no agent can reason effectively about a business it doesn't understand — what your data means, how your teams work, how your company defines "revenue" compared to the rest of the world.

**Key insight:** When every organization has access to the same intelligence, context becomes the differentiator. The enterprise that best articulates its own knowledge — its data, its processes, its meaning — will build AI that's most useful to its people.

"We built a revenue analysis agent and it couldn't answer one question. We started to realize we were missing this translation layer. We had no way to interpret human language against the structure of the data." — Joe DosSantos, VP, Enterprise Data & Analytics, Workday (video: Workday: Context as Culture).

## The AI Context Gap: one question for AI, three kinds of context

Through our work with enterprises, we've found that even a simple agent task requires three kinds of context — knowledge, expertise, and norms — working together.

**Talk-to-Data Agent** — "Why is drive-through time up this week?"

| Context type | Question it raises | Answer it needs |
|---|---|---|
| Knowledge | What does "drive-through time" mean? | **avg_dt_secs**, not the POS number Finance uses |
| Expertise | How do you investigate this? | Check **seasonality** and nearby launches before naming a cause |
| Norms | Who's asking, and what can they see? | Store manager sees **their store only**; VP Ops sees the full chain |

**Operational Agent** — "I was charged twice for my order! I need a refund now!"

| Context type | Question it raises | Answer it needs |
|---|---|---|
| Knowledge | What does "charged twice" mean here? | Could be a **duplicate**, or a hold sitting beside the real charge |
| Expertise | How do you resolve this? | Verify the **charge type** first — hold vs. double debit — before resolving |
| Norms | What's the agent allowed to do? | Resolve up to **$50** alone; above that needs a supervisor |

## Why customers love Atlan: the only proven way to create context

Customer videos:

- DigiKey: Context Readiness — Sridher Arumugham
- CME Group: Context at Speed — Kiran Panja
- Mastercard: Context by Design — Andrew Reiskind
- Virgin Media O2: Context for All — Mauro Flores

## The Context Pipeline: context doesn't come from a prompt, it comes from a pipeline

What if every agent knew what your best analyst knows? Your business systems, data estate, and people already hold the context you need. The context pipeline makes it usable.

### 1. Unify — business systems in the Enterprise Data Graph

80+ connectors pull context across your entire data estate — warehouse SQL, BI definitions, and business applications — into one living graph. That graph is what everything else in the pipeline builds on.

Capabilities: Catalog, Governance, Lineage, Quality, Glossary.

"Within the first year after that we cataloged over 18 million assets, defined more than 1300 glossary terms. Atlan had lineage across our on-prem Oracle databases, BigQuery, and Looker." — Kiran Panja, Managing Director, Cloud & Data Engineering, CME Group

### 2. Bootstrap — let AI bootstrap your context layer

Atlan's AI agents read the Enterprise Data Graph — your SQL query history, BI semantics, and pipeline code — and generate asset descriptions, link business terms, and surface your top business questions. The first 80% of your context layer is ready before a human reviews a single line.

Capabilities: Description Generator, Term Linkage, Metrics Generator, Semantic Views, Ontology Generator.

"We're scaling context development as much as possible, and where can we leverage Atlan AI to build the most robust definitions across our data estate." — Takashi Ueki, Head of Enterprise Data & Analytics, Elastic

### 3. Collaborate — humans resolve, annotate, and certify before context ships

The AI draft is a starting point, not the final word. Your domain experts resolve conflicts between sources, annotate edge cases, and certify what's production-ready. What ships is what your team trusts.

Capabilities: Conflict Resolution, Annotation, Labelling, Certification, Feedback Loops.

"Atlan gives us a UI that our community can use to edit, update and manage classifications as well as other metadata enrichments into a verified state." — Sherri Adame, Enterprise Data Governance Leader, General Motors

### 4. Activate — certified context flows to every AI agent across your stack

Production-ready context serves every downstream tool through SQL, APIs, and the Atlan MCP server. Evals, traces, and memory feed back into the pipeline and context gets sharper with every interaction.

Capabilities: MCP Server, SQL, APIs, SDK, Evals & Traces.

"All of the work that we did to get to a shared language amongst people at Workday can be leveraged by AI via Atlan's MCP server." — Joe DosSantos, VP, Enterprise Data & Analytics, Workday

## Industry recognition: a leader across every context category

- **G2:** 95% of G2 users see Atlan as a true partner. Summer 2025 badges: Grid Leader, Momentum Leader, Grid Leader Enterprise, Easiest to Use, Best Relationship, Europe Regional Leader, Europe High Performer. [G2 report](https://atlan.com/g2-leader/)
- **Leader in the 2025 Gartner Magic Quadrant for Metadata Management Solutions.** Quoted: "The Metadata Lakehouse forms the core foundation, built on an open and highly performant architecture. It is designed to be Iceberg-native and includes a knowledge graph for business domains, vector storage, and analytics, which is purpose-built for AI." [Gartner MQ report](https://atlan.com/gartner-magic-quadrant-metadata-management-solutions-2025/)
- **Leader in the 2026 Gartner Magic Quadrant for Data & Analytics Governance.** Quoted: "Atlan stands out in AI-native governance through context-based partnerships, agentic stewardship and orchestration of enterprise agentic systems. They take a partnership and co-innovation based approach, which is reflected in their App Framework as a marketplace for context." [Gartner D&A report](https://atlan.com/gartner-magic-quadrant-data-governance-2026/)
- **A Leader and a Customer Favourite in the Forrester Wave**, Data & Analytics Governance Solutions and Enterprise Data Catalogs: Forrester Wave Leader 2024 ([report](https://atlan.com/forrester-wave-2024/)), Forrester Wave Leader 2025 and Customer Favorite 2025 ([report](https://atlan.com/forrester-wave/)).

## What we believe: context will make AI worthy of humanity's most important moments

We hold strong convictions about how the context layer should be built. These shape every decision we make.

- **Context is a Team Sport.** Your frontline teams — not just engineers — should be able to read, question, and improve the context that shapes how AI behaves. The best context comes from people working together.
- **AI-Native, Built for Change.** Your context layer should outlive any single technology cycle. Today it powers MCP and A2A. Tomorrow, whatever protocol comes next — no migrations, no rebuilds.
- **Open & Portable.** Your context should move freely across agents, models, and clouds. You should never be locked into a single vendor's representation of your own knowledge.

## Frequently asked questions

### What is Atlan?

Atlan is the context layer for enterprise AI. It sits between your business systems and your AI agents, connecting lineage from data pipelines, business definitions from BI tools and SQL logic, knowledge from SOPs, quality scores, and access policies into a unified context store. Every agent and analyst queries that context store directly — no manual context-building per use case. Gartner named Atlan a Leader in the 2025 Metadata Management and 2026 Data and Analytics Governance Magic Quadrants. Forrester did the same in its 2024 Enterprise Data Catalogs and 2025 Data Governance Solutions Waves. The only platform recognized across all four.

### What does Atlan do for enterprise AI?

Atlan gives every AI agent the enterprise context it needs: the business definitions behind column names, the lineage behind every output, and the access policies behind every query. Without this, agents hallucinate, misclassify sensitive records, or return answers compliance teams reject. Every AI output is traceable — every answer points back to the data, the definition, and who certified it.

### What is an enterprise context layer?

An enterprise context layer sits between your business systems and your AI stack. It unifies context from across the business — lineage, semantic definitions, SOPs, access controls, usage patterns — into a single graph that agents and analysts query in real time. Without one, every new agent deployment starts with months of manual context-building. With one, every new agent inherits the organization's full institutional memory on day one.

### How does the context pipeline work?

Four stages: unify, enrich, certify, activate. Atlan unifies metadata from native connectors — data warehouses, BI tools, pipeline orchestrators like dbt and Airflow. Context Agents auto-generate descriptions, metrics, and business ontology across the full data graph. Human experts review and certify — human-on-the-loop, not out of the loop. Certified context activates to every agent and tool via MCP, SQL, and open APIs. Evals and traces feed back in with each cycle, so context quality compounds over time.

### How does Atlan work with AI agents?

AI agents get enterprise context through Atlan's MCP server, SQL interface, and open APIs. A query returns the data graph, business definitions, lineage, and access policies for that specific task. Context repos version and package this knowledge, so every new agent starts with the organization's full institutional memory instead of a blank slate. No context hardcoded per use case. No starting over.

### Which enterprise systems does Atlan connect to?

Atlan connects natively to 80+ enterprise systems: Snowflake, Databricks, BigQuery, Redshift, dbt, Airflow, Tableau, Looker, Power BI, and Postgres, among others. Once connected, lineage, query history, BI semantics, tags, and quality signals flow in automatically through scheduled and event-based workflows — no manual mapping required. Atlan also layers on top of existing catalogs like Microsoft Purview and Snowflake Horizon, pulling their metadata into a unified context layer.

### Who uses Atlan?

Atlan is deployed at enterprises including General Motors, Workday, Nasdaq, Mastercard, and Virgin Media O2. AI leaders use it to give agents governed access to enterprise context. Data engineers automate lineage and discovery. Governance teams enforce policies at the asset level. AI platform teams build and deploy agents faster because business logic is already in the context layer — not scattered across prompt files and wikis.

### What analyst recognition has Atlan received?

Atlan is the only platform named a Leader in all four major analyst evaluations for metadata and data governance: Gartner's 2025 Metadata Management Magic Quadrant, Gartner's 2026 Data and Analytics Governance Magic Quadrant, Forrester's 2024 Enterprise Data Catalogs Wave, and Forrester's 2025 Data Governance Solutions Wave. No other platform has been recognized across all four.

### How does Atlan work alongside my existing data tools?

Atlan layers on top of your existing data stack. Many enterprises run Atlan alongside Microsoft Purview or Snowflake Horizon or Databricks Unity Catalog — pulling metadata from all into a unified context layer rather than rebuilding from scratch. Built on open APIs and Iceberg-native formats, context stored in Atlan stays portable: it is not locked to any vendor's proprietary schema. Switch AI frameworks, add new systems, or consolidate tools — the context layer moves with you.

### How does Atlan approach context engineering?

Context engineering is the practice of selecting, structuring, and delivering the specific knowledge an AI agent needs at each step of a task. Most teams do this manually for each agent — months of work, duplicated across every use case. Atlan automates it: context from 80+ systems is unified, Context Agents auto-generate descriptions, metrics, and ontology across the full data graph, human experts certify, and certified context activates via MCP, SQL, and APIs. In April 2026, Context Agents generated 690K+ descriptions across 50+ enterprise customers — 87% rated on par or better than human writing. Every eval and trace feeds back in. Context quality compounds with each cycle.

### How do teams get started with Atlan?

Start with a Context Workshop: Atlan's team maps your data and AI architecture, designs a context layer for a priority use case, and sets a measurable baseline. From there, a four-week Context Sprint delivers a working agent and accuracy results you can compare directly against your current approach. Most teams see the first value in weeks.

## Bridge the context gap. Ship AI that works.

[Book a Demo](https://atlan.com/forms/talk-to-sales-contact/) or [see how it works](https://atlan.com/context-layer-demo/).