Blog·Engineering

Unified Memory That Works Where You Work: Your Second Brain With Supermemory

Hi everyone, I’m Dhravya, the founder of Supermemory. I want to start with a little story behind why this product means so much to me. You can also skip straight to what it is and how it works below.

By Dhravya Shah·5 min read

Supermemory banner reading "Unified Memory That Works Wherever You Work" with a laptop and phone linked to a hub device

Hi everyone, I’m Dhravya, the founder of Supermemory.

I want to start with a little story behind why this product means so much to me. You can also skip straight to what it is and how it works below.

Anyways, when I started Supermemory one year ago, we were a consumer app that acted like a second brain. You could add your memories, documents, and notes, and chat with them.

We had a big community and a shit ton of users, but, frankly, our product’s retention was terrible. People stopped using it, and there were still several technological challenges I had to figure out.

So, we pivoted to B2B and set our heads down for one year.

Now, we have a lot of businesses using Supermemory to build memory into their apps, and in the process, we figured out some of the hardest underlying technological problems. You can look at our releases of the MCP, llm-bridge, infinite chat, and more.

I was finally starting to see AI memory in the form I imagined it to be.

Although watching the revenue graph on my Stripe dashboard go berserk is very fun, I felt it was time to go after our initial vision: a memory engine that automatically (pay attention to this) stores information from all your apps, like Google Drive, Notion, Claude, ChatGPT, etc., organizes them into sexy graphs, and allows you to chat with them.

This launch has been a whole year in the making. And, I think the Dhravya of one year ago would be proud as hell.

This kind of memory has always been my real mission: bringing effortless knowledge capture to everyone, changing the way we live, work, and think. Selling B2B licenses is a noble pursuit, lol, but this is what I live for.

And today, I’m beyond excited to finally fulfill that promise.

Thanks for sticking around. I can’t wait for you to try it.

Best,

Dhravya

What is unified memory, and how does it work?

Supermemory dashboard with an empty graph view, a sidebar menu, and a legend explaining node and edge types
The app's dashboard

When I first launched Supermemory a year ago, the biggest flaw and the reason we saw so many people stop using it was that it relied entirely on manual input.

You literally had to open our app, manually add memories or documents, and then actively chat with them later.

Over the past year, I realized something crucial: memory only works when it’s automatic. When it lives quietly in the background, seamlessly integrated into the way you already work.

Memory only works when it’s automatic. When it lives quietly in the background, seamlessly integrated into the way you already work.

Here’s what you can expect from our app:

  • Automatic, real-time memory updates: The app now lives where your AI does. It automatically captures and updates your interactions from ChatGPT, Claude, Cursor, or any other AI client supporting MCP in real-time.
  • Effortless document integration: Connect seamlessly to Google Drive, Notion, and OneDrive using our connectors, instantly importing your documents and making them available for immediate conversation. Learn how.
  • Intelligent knowledge graphs: Automatically create intuitive, visual graphs of your knowledge, interactions, and memories, bringing clarity to the way you think and remember.
  • Project-based organization: Easily segment your memories into different projects, securely isolating work from personal knowledge, or dividing topics clearly, however you see fit.

Unlike other apps, the Unified Memory Engine is designed to be accessible for normal everyday users like you and me, and not just developers. For instance, platforms like OpenMemory by mem0 require a complex Docker setup, familiarity with Python, and technical know-how to even get started.

Memory graph of 492 memories and 2,515 connections shown as a dense web of colored nodes with a legend
A demo graph we created

Arguably, the graph feature is one of our coolest additions. Right now, for every raw document that is added, Supermemory not only understands and extracts things from the document to search through them, but also makes connections between two or three items.

It is also capable of making implicit connections - stuff that the user never explicitly said, but Supermemory understood.

For you, the benefit is that our system is MUCH smarter than everything else out there. You can move beyond surface-level questions like ‘when did xyz happen,’ and do actual complex analysis and querying.

The app also forgets memories that aren’t used very frequently, based on how our brain works. Other memory systems don’t have that temporal understanding, and this intelligent understanding of when to overwrite and rewrite things makes for faster performance.

I also talked to the founder of spawn.co and he told me they do this at their company. They’ve been racking every interaction on Slack and documents on Notion, adding it to the company’s knowledge graph.

How to use the app?

We’re releasing the app to only a select few people right now. To sign up for the waitlist, head over to app.supermemory.ai

Once you’re in, start by manually adding memories in the form of notes, links, files, or connections to Google Drive, Notion, or OneDrive.

Add Memory dialog offering Note, Link, File, and Connect options to save content to memory

You can also install it using the MCP for automatic syncing. For Claude Desktop, follow the following steps:

  1. Open the Claude Desktop app.
  2. Go to Developer → Edit Config to open or generate the file:
  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
  1. Add the Supermemory MCP server entry under "mcpServers". For example:
"mcpServers": {
  "my-mcp": {
    "command": "npx",
    "args": [ "-y", "mcp-remote", "https://api.supermemory.ai/mcp" ]
  }
}
  1. Save the file and restart Claude Desktop. After reboot, your MCP server should be available in the UI.

Here's how it works:

Chat demo where Supermemory recalls the user's name and stores a new memory via an addMemory tool call

You can also add it to Cursor with just one click.

Pricing

The free plan allows you to create up to 10 memories and has basic search. For 500 memories, 10 connections, advanced search, and priority support, the price is $9/month for the first 100 users.

What’s next?

This launch is just the beginning. Next up, we’re working on a browser extension to make capturing memories across the web feel super natural. You can carry your memory engine with you everywhere.

Feel free to send us your feedback or any ideas that you’d like to see implemented.

If you’ve made it this far, thank you. I genuinely can’t wait to see what you build, recall, and rediscover with it. Sign up for a free account now!

  1. An update to supermemoryWe've discontinued the supermemory company brain and Nova. Everyone who was charged has been refunded, our MCP and plugins continue to run, and we're going all in on the memory engine.
  2. SMFS: making agentic retrieval 55% cheaper AND more accurateWe launched SMFS.ai (Supermemory Filesystem) a few weeks ago, with a simple bet: We can redesign the filesystem specifically for agents, with special files, structures, and commands that it can use for it's tasks. Today, SMFS is used by hundreds of companies to power their agents.
  3. Introducing Dynamic Dreaming: supermemory now connects the dots, for you.Dreaming is magical. TLDR: We're launching Dynamic Dreaming in supermemory today, which automatically works if you're using supermemory in any way - API, OpenClaw, Hermes agent, etc.
  4. Dear reader, we just made supermemory insanely cheap... the Context CloudWhen I first started building supermemory, I had one goal: To build the best memory system for AI. I would talk to customers, and find out that memory was not the only thing they needed - They were all setting up 7-8 different vendors at the same time.
  5. Introducing @supermemory/tools v2.0.0Today we're releasing v2.0.0. This release unifies the API across all agents sdk integrations from AI SDK to Mastra, makes conversation identity a first-class concept, and ships with memory saving on by default.
  6. supermemory will make your Hermes-agent crazy powerfulToday, we are launching supermemory support to your Hermes agent TLDR: you can use supermemory now in your Hermes agent, it totally free to get started - https://supermemory.ai/docs/integrations/hermes In case you missed it: Hermes Agent is a self-improving AI agent from Nous Research.
  7. Solving the Precision-Recall Tradeoff: Search Result AggregationWhen you're building memory for AI, search is your foundational layer. The way search generally works is straightforward: the user defines a query, and then sets a limit (top-K) on how many search results they want returned. Usually, this is set to 10 or 20.
  8. Infinitely running stateful coding agentsWe built a plugin for Claude Code and OpenCode that gives your coding agent persistent memory. It remembers your preferences, learns your codebase, and never loses context mid-conversation. The result is an agent you can run for months without starting over.
  9. Why everyone is complaining about OpenClaw's memory (it sucks) - and why supermemory fixes it.TLDR: Today, we are releasing a new version of our openclaw plugin - https://github.com/supermemoryai/openclaw-supermemory. This post is going to be a bit technical, so bear with me (or bookmark for later!) In this post, I will talk about what we do about OpenClaw memory, and how we fix it.
  10. We added supermemory to Claude Code. It's INSANELY powerful now...Today, we are launching the Supermemory plugin for Claude Code! TLDR: You can use supermemory in claude code now. - https://github.com/supermemoryai/claude-supermemory Claude code has genuinely changed how I work. But there's this one thing that drives me crazy...
  11. Clawd / Molt bot's memory SUCKS. We gave it supermemory.I'm the founder of supermemory. Clawd/Molt bot is blowing up right now, with many, many use cases. I set it up, too, and have been using it through telegram. TLDR: just go to https://supermemory.ai/docs/integrations/clawdbot to set up supermemory for your clawd bot.
  12. Catch up with our UNFORGETTABLE Launch WeekOver the last year, one belief has guided almost everything we’ve built at Supermemory AI becomes meaningfully useful only when it remembers. Memory shouldn’t be something developers rebuild from scratch. It shouldn’t be fragile, expensive, or trapped inside a single tool.
  13. Empowering the Next Generation of Founders: Supermemory Startup ProgramIf there’s one thing we’ve learned while building Supermemory, it’s that most startups don’t fail because they didn't build features; they fail when infrastructure slows them down, or they built too slow.
  14. Building code-chunk: AST Aware Code ChunkingAt Supermemory, we're building context engineering infrastructure for AI. A huge part of that is dealing with code: ingesting repos, understanding structure, and making it searchable. The problem is that most code chunking solutions are terrible. We built code-chunk to fix this.
  15. Supermemory raises $3 million with the best memory engine for LLMsToday, I am excited to announce our first funding round to accelerate our mission of building an interoperable, scalable and reliable memory for LLMs and agents. Memory is one of the hardest challenges in AI right now.
  16. Supermemory just got faster on PlanetScaleWhat is Supermemory? Supermemory completes the missing part of the LLM puzzle: memory. Just as memory is crucial for human intelligence, it's essential for truly intelligent AI systems.
  17. Faster, smarter, reliable infinite chat: Supermemory IS context engineering.People are obsessed with prompts and prompt engineering. Sure, what you say is important, but what the model knows when you say it is the difference between a stateless text generator and an intelligent AI system. In short, context is the most crucial component.
  18. We solved AI API interoperabilityOne API to rule them all, One spec to find them, One library to bring them all and in the TypeScript, bind them. When we were building the the Infinite Chat API, initially, we only supported the OpenAI format. This was fine, until a lot of our customers started asking, asking for more.
  19. The UX and technicalities of awesome MCPsLast month, we launched the Supermemory MCP, mostly to test our own infrastructure and get some initial traction. It blew up. To my absolute surprise, the initial launch itself got half a million impressions (!!!). Then, we launched and got #2 on ProductHunt too.
  20. Architecting a memory engine inspired by the human brainLanguage is at the heart of intelligence, but what truly powers meaningful interaction is memory — the ability to accumulate, recall, and contextualize information over time. Large Language Models (LLMs) have mastered language, but memory remains their Achilles’ heel.