Skip to content

lasect/ambala

Repository files navigation

ambala

Contextual search engine built on graph + vector retrieval.

Overview

ambala is a hybrid retrieval system that combines semantic similarity with graph traversal to produce context-aware results.

Instead of treating search as a flat nearest-neighbor problem, Ambala evaluates relevance within a structured graph.

This project focuses on:

  • Combining vector similarity with graph traversal
  • Supporting context-constrained queries
  • Making retrieval explainable

Status

Early stage. Core API routes and query layer are being implemented.

Architecture

  • HelixDB → graph + vector storage
  • TanStack Start → API layer
  • Vercel → deployment
  • Gemini → embeddings

License

MIT

About

contextual search engine powered by helixdb

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors