Launched this week
Trisynapse
Turn your scattered sources into a digital brain
7 followers
Turn your scattered sources into a digital brain
7 followers
Trisynapse is a local-first, BYOK AI research workspace. Turn scattered papers, PDFs, bookmarks, notes & videos into a private digital brain. The app auto-summarizes, maps concept graphs, and compiles sources into auto-updating wikis. Work seamlessly with a built-in split-view Browser, local retrieval Chat, AI Assistants, and infinite Canvas whiteboards for note-taking. 100% offline-first (except AI API calls) and private: your vault stays on your machine
Hey Product Hunt 👋, Shanmukh here, developer of Trisynapse.
To be honest, I built this because I was losing my mind trying to keep up with the new AI and LLM research. My bookmarks, Twitter saves, and browser tabs became bloated with unread sources. I tried keeping them in Notion and Obsidian, but manually organizing everything felt like a slow process.
So, I built Trisynapse to be the tool I actually wanted to use. It's a completely local-first, Bring-Your-Own-Keys (BYOK) desktop workspace.
Here is how it works:
- Builds a private digital brain: You point the app to a folder on your computer (your "vault"), and everything - your PDFs, text extractions, vector embeddings, and concept maps stays right there. No accounts, no subscriptions, no cloud sync.
- Does the reading and synthesizing for you: Once you add files or links, it auto-extracts the clean text, maps connections, and compiles everything into interlinked, self-updating Markdown wiki pages.
- Scans for updates in the background: You can set up keyword alerts, and the app will quietly scan Arxiv, Tavily, and X in the background to surface new discoveries.
- Keeps you in one place: Instead of jumping between ten apps, you have a split-screen browser (with sidebar chat), a floating AI assistant, and an infinite whiteboard canvas (powered by Excalidraw) for sketching out notes.
Since this is a fresh MVP, I would love to know: Does this workflow fit into how you read or research?
Additionally, as I look to scale the vision for Trisynapse, I am currently looking for investments and potential co-founders who want to take Trisynapse further. If you or anyone in your network might be interested in collaboration, I would love to chat further.
Ask questions, give feedback, and suggestions!!
Best,
Shanmukh
how does the concept graph actually work when you throw in a messy mix of PDFs, video transcripts, and random notes — does it figure out connections on its own or do you have to manually link things first?
@asmin467991 Trisynapse automatically builds the concept graph without requiring manual linking. When you add messy PDFs, video transcripts, or notes, the LLM automatically extracts key concepts. These are then mapped to their concepts and automatically connect overlapping concepts based on co-occurrence in the same source (weighted by frequency) or explicit AI-detected relationships. The entire semantic web is generated dynamically and automatically.
Finally tried it for an hour with some research papers and the concept graph actually connected ideas I had not linked in my own notes. The split-view browser next to the canvas felt really natural for working through sources.
@efeerkolpgav Thank you. Have you tried any Chat Interfaces available? Wiki Chat, Main Chat, Browser chat?