Hi HN – founder here.
We built Telio to power AI agents for 24/7 call and text support.
Voice calls and texts need fast access to context. Our approach: aggregate data from multiple sources into a unified data lakehouse using APIs, webhooks, and logical replication.
Key advantages:
- Sandboxed read-only environments for AI agents, so they can't disrupt other systems
- Storing and processing large data volumes on S3 at a fraction of typical costs
- Retrieving all context at once, reducing tool calls and LLM token usage
- Building and querying vector embeddings for semantic search
- Granular column-level permissions to manage PII access
We query the lakehouse as if it were PostgreSQL, but with much better performance on complex queries using our open-source project https://github.com/BemiHQ/BemiDB.
We’d love feedback from the HN community!