​
Getting Started

​
What is Prometheux?

Prometheux is an ontology-native data processing engine. It allows you to define ontologies that are executable—meaning they can be used directly to query and process data across multiple systems. An ontology in Prometheux is not just documentation. It is a runnable model that defines:
  • Entities and relationships
  • Semantics and constraints
  • Query and processing logic
Once defined, the same ontology can be used consistently across analytics, applications, and AI systems.

​
How Prometheux Works

Prometheux introduces an ontology layer that sits above your existing data systems. Instead of moving or transforming data into a single location, Prometheux:
  • Connects to data where it already lives
  • Applies ontology-defined logic at query or execution time
  • Produces results with full lineage and traceability
This allows teams to work across databases, warehouses, and platforms without requiring mandatory migrations or ETL pipelines. Prometheux architecture Figure: Prometheux architecture.

​
Key Concepts

​
Executable Ontologies

Ontologies in Prometheux are directly executable. They can be queried, composed, and used to drive data processing workflows.

​
Ontology-Native Queries

Queries are expressed in terms of ontology concepts rather than physical schemas, reducing coupling to underlying data structures.

​
Distributed Data Access

Prometheux can operate across multiple data sources simultaneously, resolving semantics at runtime.

​
Lineage and Traceability

All results retain explicit links back to the ontology definitions and underlying data sources.

​
Supported Environments

Prometheux is designed to run in modern data stacks and can be deployed alongside existing platforms.
  • Native integrations with Databricks and Snowflake
  • Deployable in cloud or on-prem environments
  • Works with structured data across multiple storage and compute systems

​
Who This Documentation Is For

This documentation is intended for:
  • Data engineers integrating Prometheux into existing platforms
  • Analytics and BI teams querying ontology-defined data
  • AI and application developers requiring consistent semantics
  • Architects designing ontology-driven data systems
No prior ontology tooling is required, but familiarity with data modeling concepts is recommended.

​
What’s Next

Start by installing Prometheux in the environment that fits your team:
  • Installation — Choose between Cloud, Docker, Snowflake, Databricks, or on-premise
Then explore the fundamentals:
  • Vadalog Reference — The language underpinning the Prometheux reasoning engine
  • Python SDK — Connect to Prometheux programmatically