BM25 + vectors + RRF Local index Cited answers

Find answers in your documents — with proof

Holusight indexes PDFs, Office files, markdown, and code on your infrastructure. Hybrid retrieval fuses keyword and semantic search; every result links to the exact source chunk.

Start a pilot Quick start

The problem

Teams store policies, contracts, runbooks, and code across folders and wikis. Generic chatbots hallucinate. grep misses concepts. Holusight is built for retrieval you can verify — ranked chunks with file, page, or line citations.

What ships today

Hybrid search (local)

SQLite FTS5 BM25 + LanceDB vectors merged with reciprocal rank fusion. No LLM required for search.

Ask with sources

Optional LLM synthesis (Claude, Azure OpenAI, OpenAI, or Ollama). Retrieved chunks are always shown alongside the answer.

Docker pilot deployment

FastAPI server + browser UI, API key auth, read-only document mount, persistent index volume.

CLI & Python API

python -m codesight index|search|ask and the CodeSight class for automation.

This website is static. holusight.com does not index or process your files. The paid product runs on your VM or cloud account.

How it works

See the citation contract for how file types map to page vs line metadata.

Honest fit

Holusight is a focused consulting product — not a Sourcegraph or Glean replacement.

Good fitNot yet
Single-team pilot over a document folderCompany-wide SSO
Cited search across PDFs + codeLive M365 / SharePoint sync
On-prem or customer-cloud VMMulti-tenant SaaS on holusight.com
Hybrid BM25 + semantic retrievalGraphify code-graph integration

Local-first boundary

Stays local

Indexing, embeddings (default model), BM25, vector search, and the index files on your disk or volume.

May leave your network

Only when you configure Ask with a cloud LLM, or opt into Voyage/OpenAI embeddings. Use Ollama for fully local answers.