myNeutron - Portable memory that follows you across AI chats.
by•
Lose context between AI chats and LLMs? myNeutron gives you portable AI memory: capture from the web, Gmail, and Drive, search semantically, then drop the right context into any model in one click or through our MCP.
Replies
👋 Hey Product Hunt!
I’m Burhan, I build and use AI every day, and the one thing that kept slowing me down was that I was locked in on one AI.. The context never followed me
• Research lived in tabs and inboxes
• Every new ChatGPT/Claude/Gemini chat meant re-explaining
• Copy-paste chains broke, answers got inconsistent
So we built myNeutron, a simple way to make context portable.
What it is
myNeutron is a portable AI memory layer. You save information once and reuse it in any AI chat.
How it works
📥 Capture pages with the Chrome extension; optionally auto-save Gmail & Drive into small context packets called Seeds
🔎 Find & combine Seeds with semantic search so you bring only what’s relevant
🔌 Inject that context into ChatGPT/Claude/Gemini in one click (extension) or via MCP
🧠 Or ask our AI Assistant to answer strictly from selected Seeds, with citations
🔒 Control & privacy: you choose what to save/export; originals stay where they are.
What’s live today
• Chrome capture, Seeds, semantic search, combine, one-click paste, AI Assistant with citations and MCP integration
• Rolling out in the coming weeks: deeper Gmail/Drive automations
Early tester perks
During beta, you’ll get an extended free tier and a direct line to shape the roadmap.
We’d love your feedback
• What feels clunky in the capture → context → answer flow?
• Which connectors matter most for you?
• What would make myNeutron a true drop-in to your workflow?
If you try one thing: save your first 3 Seeds, combine them, and paste into any model, then ask the same question across tools. Tell us where it breaks and we’ll fix it fast.
UI Bakery
Love the idea of portable memory across AI chats. Do you have thoughts on how myNeutron handles conflicting or outdated context? E.g. if a Seed says one thing and later content contradicts it?
@vladimir_lugovsky Hey Vladimir, great question! The conflict resolution is actually one of the key design considerations.
For manual context generation, you actually have control - when generating context, you can select the exact Seed you want to update. So if information changes, you're explicitly choosing which Seed gets the new context, avoiding unintended conflicts.
For MCP and AI assistant usage (where context flows more automatically), that's where the conflict handling becomes more important:
Timestamp priority - More recent information generally takes precedence, but the system flags when there's a conflict so users are aware both versions exist
Explicit updates - Users can explicitly update or deprecate Seeds when information changes, rather than having old context silently override new learning
Conflict notifications - The AI assistant gets notified about contradictions and can surface them in conversation ("I have conflicting information about X - here's what I knew before vs. what you just told me")
Seed versioning - Seeds can be versioned so you maintain a history of how information evolved rather than losing context entirely
The goal is to make context portable and maintainable - so you're not locked into outdated information, but you also don't lose the thread of how your knowledge evolved over time.
Does that address your concern, or would you want to see different behavior?
Great idea really, and solve lots of problems, and you need to wide your range to consider other browsers as well...
myNeutron is such a clever solution—portable AI memory can really help maintain context across chats and tools. Do you plan to support integrations with more AI platforms or apps in the future?
Huge respect to the team for building something this smooth.
Memory is the missing layer AI always needed — and Neutron nailed it
.
myNeutron finally gives AI a real memory. It’s like having Notion, ChatGPT, and Google Drive fused into one everything you save becomes searchable and contextual.
I’ve been using it for research, and it remembers exactly what I need. This is what AI productivity should feel like smart, personal, and unforgettable.
It is a good way to start
Massive props to the team for delivering such a seamless experience. Memory finally feels native to AI thanks to Neutron.
One thing that stands out for me is the fact that memory goes across AI chats and I can't wait for what the team fully has in store.
I've been using neutron AI for a while now and honestly my experience has been really amazing
I love the fact that I can easily compress lots of documents into seeds and these saves me more space on my device and is less stressful to access those files