Enterprise search for teams drowning in Slack threads

What it does Gerev ingests workplace data from tools like Slack, Confluence, Jira, and Google Drive, then exposes it through a natural-language search interface aimed at help-desk teams. The goal is to surface scattered internal conversations and documents without hunting through individual apps. It is available as a managed cloud service or as a self-hosted Docker deployment with optional GPU support.

The interesting bit The README treats integrations as the primary selling point: most of the page is a community-sourced checklist of connectors, with several still in open PRs. That makes the project look less like a novel search algorithm and more like opinionated plumbing that normalizes a dozen SaaS APIs into one queryable index. A guide for adding custom data sources is included.

Key highlights

• Ships with connectors for Slack, Confluence, Jira, Google Drive, Mattermost, RocketChat, GitLab Issues, and Bookstack.
• Accepts plain-English queries such as “How to do X” rather than requiring exact keywords.
• Community-driven roadmap: Zendesk, Notion, Azure DevOps, Trello, and others are listed as in-progress or in PR.
• Self-hostable via Docker; the container can use GPU acceleration or fall back to CPU.
• Provides a documented path for adding custom internal data sources.

Caveats

• The README never explains the underlying AI mechanics—whether it uses embeddings, LLMs, or keyword ranking—so what “AI-powered” means in practice is unclear.
• A significant portion of the integration checklist is still pending, meaning the value depends on whether your stack matches the finished connectors.
• All managed-cloud and demo links point to app.klu.so, a domain the README does not explain or acknowledge.

Verdict Worth evaluating if your internal support team wastes hours hunting through Slack and Confluence. Look elsewhere if your org lives in Microsoft Teams, Notion, or SharePoint—those connectors are still on the todo list.