From REST to reasoning: ingest, index, and query with dlt and Cognee

​In this hands-on workshop, we'll walk through building a knowledge graph from REST API documentation using dlt and Cognee. Rather than treating APIs as flat data sources, we'll model their underlying concepts—resources, relationships, endpoints, and behaviors—as a structured graph of interconnected nodes.

​Using dlt, we'll extract and transform documentation from multiple REST APIs into structured representations. These are then embedded into a shared semantic space using Cognee, where a central ontology node anchors all inputs, linking method definitions, OpenAPI specs, and DLT-specific constructs across APIs.

​We'll explore:

• ​How to extract and normalize API docs into semantic node sets

• ​How to build and extend a central ontology that connects multiple API sources

• ​How to visualize and query the resulting knowledge graph to answer conceptual questions.

​By the end of the session, you'll know how to construct and query a multi-source, semantically-indexed knowledge graph, enabling a richer understanding of complex API ecosystems.

​About the speaker:

​Hiba is an AI Engineer at dltHub. She works with data pipelines, data modelling, LLMs, and ML tasks in her current role. Before that, she had worked at other startups in data roles.

​​This event is sponsored by dlthub.

​​​​​DataTalks.Club is the place to talk about data. Join our Slack community!