SEARCHSearch & retrieval
Jev-Powered Knowledge Base
Cachetronaut@cachetronaut_
This project builds a local-first database that uses multiple search criteria and Jev to locate and retrieve relevant information, creating a continuously learning feedback loop.
One of the many useful use cases for @typesafeai's Jev is building a knowledge base. Non‑frontier‑level models often struggle to locate the right content and context.
To solve this, I created a small, local‑first database that:
- Combines several search criteria to pinpoint the correct information.
- Falls back to Jev when uncertainty remains; Jev returns a matching result.
- After approval, reuses the same key phrases and criteria for the next search.
This process forms a continuously learning, growing feedback loop.
Related builds
Vectorless RAG with Jev
This project introduces a novel approach to Retrieval Augmented Generation (RAG) that replaces traditional embedding methods with a structured, table-of-contents-based navigation system, significantly improving speed and reducing costs by offloading non-generative tasks to Jev.
