Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Airweave: Indexing Apps for Agents
Learn how Airweave automates ingestion, transformation, and indexing of apps, APIs, and databases, handling authentication and syncing to vector and graph stores for AI agents.
In this talk, we’ll show how we built Airweave, an open-source tool that makes source app, API, and database searchable for AI agents. Airweave automates ingestion, transformation, and indexing, while handling authentication, mapping, interweaving, and syncing to vector stores and graph databases. Originally built to solve retrieval challenges in our own AI agent stack, we decided to open-source it. We’ll walk through its architecture, the challenges re ran into when building it (as well as the ones we’re facing right now), and structured data retrieval for AI agents.
Airweave enables agent semantic search using FastAPI, PostgreSQL, and Qdrant vector database.