Memvid retweeted
This year has been an interesting one for me. Between contributing to the Rust compiler and Git, joining Super30 3.0 by @kirat_tw, and now starting as a Founding Engineer at @memvidAI less than two months into Super30, I have been fortunate to work on some amazing open source projects.
Memvid is a portable, serverless memory layer that gives AI agents persistent memory and fast recall. It's model-agnostic, multimodal, works fully offline, and is being used across a wide range of real-world AI applications.
Instead of managing vector databases and complex retrieval infrastructure, Memvid packages your data, embeddings, search structure, and metadata into a single portable file that can be shared, moved, and queried anywhere.
@memvidAI isn't limited to text. It also supports multimodal memory, allowing developers to embed and retrieve images alongside text from the same memory file.
Even more interesting is the architecture behind it. Rather than drawing inspiration from traditional databases, @memvidAI borrows ideas from video encoding by organizing memory as an append-only sequence of Smart Frames. This enables:
- Append-only writes without modifying existing data.
- Queries over historical memory states.
- Timeline-style inspection of how knowledge evolves.
- Crash safety through immutable committed frames.
- Efficient compression inspired by video encoding.
From long-running AI agents and codebase understanding to enterprise knowledge bases, Memvid makes persistent AI memory simpler to build, deploy, and share.
Written in Rust with sdk support for Node.js, Python and Rust, Memvid delivers blazing-fast retrieval with the memory safety and performance Rust is known for.
@memvidAI already has 15.8k+ GitHub(github.com/memvid/memvid) stars, and we're just getting started.
If you're interested in Rust, AI infrastructure, or AI agents, give Memvid a try and let us know what you build.
We're investing heavily in the open-source community with engineering deep dives, contributor spotlights, and a welcoming environment for both new and experienced contributors.
Join our Discord(discord.gg/vHvRHhJnHC), share your projects, open an issue, or send a PR, we'd love to build @memvidAI together.
Thanks to @its_mohamedomar and @saleban for easy onboarding and @kirat_tw for the referal.
Let's build interesting stuffs together with AI.
xitter.cf/usman_akinyemi/status/…
Locked In and grinding hard for the next Six months at Super30 3.0.
The first guy from another continent I guessed.
@100xDevs @100xSchool @kirat_tw
Memvid retweeted
Developer building an 'AI digital twin for jobs' chose @memvidAI over Pinecone/Weaviate.
He's documenting his journey in a 4-part blog series here: 👇 Good read
medium.com/@schwicht/part-2-…
Memvid retweeted
Claude-brain by @memvidAI 👇
Memvid retweeted
Finally got tired of scrolling through 5000 screenshots named "Screenshot 2012-01-15 at 10.32.41.png"
Made a thing: github.com/memvid/screenshot…
It OCRs all your screenshots so you can search by the text in them. also has local AI vision for photos (uses ollama) so you can search "that stripe error" or "guy with headphones" and
It actually works. no cloud, runs locally.
Memvid retweeted
Replying to @its_mohamedomar @memvidAI
honestly this is crazy cool man. zero infra and 5 min setup is insane. been following memvid since the start and this v2 looks even more insane
Memvid retweeted
🚀 Memvid is #1 GitHub Repo of the Day -> 2nd day in a row!
Couldn’t have done it without this incredible team and an amazing dev community.
Thank you for the love ❤️We are just getting started.
👉 github.com/memvid/memvid
Memvid retweeted
We just killed every RAG and Vector Database company
(by storing AI memory in video frames)
- Zero infrastructure
- 5 minute setup
- Hybrid search
- Sub-5ms retrieval
- Fully portable
- Open Source
Introducing @memvidAI v2
Memvid retweeted
Big news in the Memvid journey.
Over the last year, @its_mohamedomar and I went through multiple failures, hard pivots, and plenty of moments where the easiest move would’ve been to throw in the towel.
But we kept building. Kept shipping. And eventually… something clicked.
A few months ago, Memvid exploded:
10M+ views, over 10k GitHub stars, thousands of users, and a waitlist growing fast.
Seeing people fork, remix, and push the product in new directions has been unreal.
That momentum helped us close our round. And last week, we made a big move, literally.
We relocated to Chattanooga, Tennessee to join Brickyard.
No distractions. No shortcuts. Just brutal clarity and execution.
We’re working on a massive v2 right now, and I can’t wait for people to get their hands on it. 👨🏾💻
More soon! 🚀 @BrickyardVC
Memvid retweeted
Huge update!
@memvidAI just entered its next chapter.
After multiple failures, long nights and some hard pivots... something finally clicked. 👇














