The memory and context cloud for AI agents
Get your company brain now - https://t.co/X2cl3JbHsB
https://t.co/9JWFW4ca89git.new/memory San Francisco, CAJoined May 2024
People hear "AI dreaming" and picture some conscious sci-fi thing. It's way more boring than that and way more important.
@_prasanna_ap explains it in this clip: dreaming is just your memory cleaning itself up in the background. re-reading what it knows, connecting the dots, dropping what's stale. the model doesn't change, your memory gets sharper.
and that's the quiet engine behind the bigger thing he's betting on: a personal brain that's actually yours!
One you own and control, that grows with you, and eventually knows you well enough to help you the way you would.
full clip below.
its my birthday! I turned 21
all I want is everyone to use @supermemory 🫡
We're giving 50% OFF ALL PLANS for everyone who signs up today!
Here's some ways to use supermemory, and link for the discount ⬇️
We just shipped the supermemory showcase.
Coding agents, voice assistants, sales tools, games, research bots, personal apps, all built on supermemory, all in one place.
Made something on @supermemory? drop your project and we'll feature it ↓
We just shipped the supermemory showcase.
Coding agents, voice assistants, sales tools, games, research bots, personal apps, all built on supermemory, all in one place.
Made something on @supermemory? drop your project and we'll feature it ↓
Company Brain is built on some incredible infra:
∗ @threepointone - Agents SDK
∗ @CloudflareDev - Durable Object & AI Gateway
∗ @daytonaio - Secured Sandboxes
∗ @supermemory - Best Memory Engine to power your company's memory
article coming soon. My first in years, this feels like a good one to come back with.
The @supermemory company brain is super powerful
It is also insanely token efficient.
So we're decreasing its pricing to just $100/month.
You can use it with any model (Claude Fable etc) and connect it to all your tools, you can have a thousand employees. Just 100
Everyone wants AI with a bigger memory, we think that's the wrong goal..
We sat down with @sohamdaga22, AI researcher at @supermemory, for a deep dive into how machines actually remember you.
what we get into:
- why every AI forgets you
- why forgetting is the hard part, not remembering
- memory that understands time
- going past reranking with aggregation
- buckets for agent memory
- SMFS, cutting tokens by 60%
A conversation on the future of context, full episode below!
The @supermemory company brain is super powerful
It is also insanely token efficient.
So we're decreasing its pricing to just $100/month.
You can use it with any model (Claude Fable etc) and connect it to all your tools, you can have a thousand employees. Just 100
He sold a company at 16. Then raised $3M at 19 without a co-founder.
Today he gave the whole thing away.
@supermemory now runs fully local, self-contained, and the binary is open source.
Solo Founders Podcast ep 14 is live with @DhravyaShah of Supermemory.
00:49 How a side
"Memory is a lego set with no instructions. what's perfect, you decide."
@DhravyaShah on why we don't force one way to build memory.
Tool calls, hooks, MCP, hybrid retrieval, a filesystem your agent can grep, all of it works. @supermemory gives you the blocks, you build what your use case actually needs.
i built a linkedin sales dashboard that gets sharper every time i use it, instead of decaying like every other cold outreach tool.
it runs on @supermemory:
→ add a prospect once and it watches their X & company site for buying signals (funding, new hires, launches). each one saved to supermemory, tagged to them.
→ search the whole pipeline in plain english. type "EU founders hiring this week" and supermemory ranks it by meaning
→ hit draft and claude pulls that prospect's full memory to write a DM about something they actually did.
→ every DM and reply logs back, so six weeks later their whole history is just sitting there.
so every interaction makes it smarter.
comment and i'll send it over.
We call it context pollution: when your agent has so much stale, contradictory context that it confidently acts on a decision you reversed last week.
what it actually looks like:
→ stale facts it still thinks are true
→ two contradictory things and no clue which one wins
→ the detail that mattered, buried under 200k tokens of noise
→ old context it drags into every answer
a bigger window doesn't fix this but throwing the junk out does!
2K Followers 4K FollowingBuilder. 25 years across government systems, cloud, Web3, and now AI agents. Founder of @perk_os. I write about infrastructure, identity, and what comes next.
388K Followers 2K Following3x Forbes 30 U 30 🚀 GP @KOLCapital💰 600+ Investments 🤝 Adventure Capitalist 🌏 100+ countries 🏛️ Minister of IT @Liberland_org 💻 AI Leader of The Year
299 Followers 3K FollowingI have changed this so many times. God, Husband and father of 2 is the only constant. I have experience as developer, project manager and product owner.
13K Followers 117 FollowingBuilding the first full-stack AI financial infrastructure | Find our 24/7 Cafes, Events, Bus Routes: https://t.co/y3PCKlqgp8
548 Followers 166 Followingai strategist, letting ai run experiments on people —- agentic trading, innovation research @invisiblebags, building @loopyagentai
4K Followers 4K FollowingFounder/CEO of Graphlit (@graphlit): Operational Context Layer for AI Agents 🚀 @zine_ai @dossium 👋 ex-MSFT, PA born, Seattle bred. Dad to dogs/humans