Hemory
Keep listening. Searchable memory for your AI agents.
640 followers
Keep listening. Searchable memory for your AI agents.
640 followers
Hemory comes from Hear + Memory. It's not meeting transcription: Hemory listens on the phone or Apple Watch you already own, and keeps every conversation as memory. Your day is auto-split into moments with speaker labels, then settles into a private, searchable memory. Connect it to Claude, Codex, Cursor, or any agent over MCP, and your AI finally has real context to revisit what you heard and build on it.
Launch Team
CloudflareStill paying to deploy your apps? Try Cloudflare Workers for free
Promoted
the phone + apple watch angle feels right, especially the idea of keeping conversations searchable instead of another pile of transcripts. curious how you handle moments people don't want remembered?
@ateeq_syed Hemory lets you set your own listening schedule, so you can keep listening turned off during the parts of your day you’d prefer to keep private.
How do you decide what's worth remembering vs. noise? In my own Claude Code setup, memory files grow fast and I rarely know which parts the agent actually reads
Brilliant concept — continuous 24h listening with auto event segmentation is a game-changer. No more manual recording. Love the MCP integration too. Upvoted!
@new_user___2572026b0e111b149ce38cb Thanks so much for the upvote and kind words! Give it a full day of listening, then ask your AI about it over MCP. That's when it really clicks.
@new_user___2572026b0e111b149ce38cb Try recording continuously for a week, then connect your AI agent to look back and summarize your conversations. You might be surprised by what you discover!
Came across Hemory today. Interesting approach.
I’ve used some AI meeting note tools before, but the problem is I rarely go back to those notes. Turning conversations into something searchable and connected over time makes more sense to me.
The speaker identification looks useful too. Will be interesting to see how well the memory works after using it for a few weeks.
@evanbuildsai Thanks for checking it out! That's the idea. You don't have to go back to your notes, your AI does it for you. Let it listen for a full day, then ask your AI over MCP about your day. Hope you love it!
The transition from "recording a meeting" to "seamlessly capturing daily memories" is such a huge design shift. Love that it automatically splits audio by context so you aren't left with an overwhelming 24-hour file, and MCP integration with tools like Claude makes it actually actionable. Definitely upvoted!
@romy_chou Thank you so much for the upvote! We spent a lot of time on splitting the day by topic. After a full day of listening, you see every conversation neatly laid out on your timeline. That's a big aha moment for us. Can't wait for you to have yours!
audio first memory for agents is smart 🎧 hows it handle overlapping speakers?
Your Apple Watch just became the intern who actually takes notes haha. Pretty wild to give agents real context instead of another empty recap. Nice work guys! @wang_yingqi @jasper_z @bushcraft