The map of
your entire business.
An ontology defines the entity types, relationships, and rules that hold in your business. Atlan mines the first draft from your lineage, SQL, and BI logic instead of starting you at a blank page, then versions and tests it like code.
Interactive diagram of five connected entities: Campaign, Customer, Order, Shipment, and Support Ticket. Use tab to move between entities, enter to read what each one means, and arrow keys or drag to reposition them.
Trusted by AI-forward enterprises
"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 & Analytics, Workday
"The output shifted from solid generic descriptions to something that felt like it had been written by someone who understands our business. That's the moment I stopped thinking about this as a time-saving tool and started seeing it as a strategic capability."
Sridher Arumugham
Chief Data Analytics Officer, DigiKey
"We were stunned and perplexed by the quality of the content. How could the agents create such high-quality context from lineage, SQL, and dbt logic? It shows how much business information is hidden in metadata that we can't see with human eyes. But agents consume it all and organize it."
Andrew Reiskind
Chief Data Officer, Mastercard
"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
WTF IS THE CONTEXT LAYER
A bi-weekly live series for AI leaders and builders. One burning question per episode, an open AMA floor, and guests who've actually built context infrastructure.
- Ontologies have existed for decades. Why are enterprises funding them now?
- What has to exist before you build an ontology?
- Can an LLM build the ontology for you?
- Is the context layer just a new name for the ontology?
One question.
Three kinds of context.
Knowledge is what a thing is and how it connects. Expertise is what follows from that. Norms are which rules apply. An ontology is where all three get declared.
Why is drive-through time up this week?
Question It Raises
Context Type
Answer It Needs
What kind of thing is this?
A ServiceMetric, recorded by a Store, which belongs to a Region — not a raw machine reading
What follows from that?
A Region rolls up its Stores, so a chain-level spike traces down to the stores under it
Which rule decides?
A Store closed for refurbishment is excluded from chain averages
No agent runs on
an ontology alone.
It says what can be true. Which data to trust, how to compute a number, and what the team decided last time all live in other structures.
What are the business’s entities, relationships, and rules?
Entity types, named relationships, and the logic that has to hold. It carries no data of its own, which is the point: it is the blueprint a knowledge graph gets filled in against.
Explore OntologyONTOLOGY
ENTITIES + RELATIONSHIPS
CLASS HIERARCHY
RULES — ALSO PART OF THE ONTOLOGY
Don’t start from a blank page.
Atlan mines the first draft from your lineage and SQL.
The entities and relationships are already in your systems. They are just not written down.
Your business already encodes its own ontology, badly and in a dozen places: in join paths analysts repeat, in the filters every production query applies, in BI semantic logic, and in glossary entries somebody started three years ago. Atlan reads those signals and proposes the entities, relationships, and definitions they imply.
ORDERS_RAWDIM_CUSTOMERSREVENUE_AGGTreated like code, because a wrong rule is a wrong answer at scale.
A draft ontology is a starting point. The rules that matter most are usually the ones no system recorded, which is why a human owner decides what becomes canonical, what it replaces, and where it propagates.
One ontology, readable by every agent you run and every agent you have not built yet.
An ontology that only one agent framework can read is a private ontology. Atlan stores context in Apache Iceberg and serves it through whichever interface a consumer speaks, so the meaning outlives the tool that happened to need it first.
Cortex AnalystSee the context layer, built live.
Learn more about ontology with Atlan.
Ontology in AI
What an ontology is in the age of AI — the formal model of your business concepts and how they relate that lets agents reason about your data instead of guessing.
Ontology-First AI Architecture
Why leading teams design AI systems around a governed model of meaning first — instead of bolting context onto agents after the fact.
Active Ontology
An ontology that stays alive — continuously updated from your data so the meaning your agents rely on never drifts from how the business actually runs.
Ontology in AI
What an ontology is in the age of AI — the formal model of your business concepts and how they relate that lets agents reason about your data instead of guessing.
Ontology-First AI Architecture
Why leading teams design AI systems around a governed model of meaning first — instead of bolting context onto agents after the fact.
Active Ontology
An ontology that stays alive — continuously updated from your data so the meaning your agents rely on never drifts from how the business actually runs.
Give your agents the map
your business runs on.
