D 🦭/acc retweeted
Every company is going to need a system of record for its agents
The length of software tasks AI agents can complete has doubled roughly every 7 months since 2019, and recent models are moving even faster, from 30 seconds of human work in 2022 to a full workday today.
Yet inside most companies, AI still isn't trusted to reliably update records in Salesforce. "AI does not have an intelligence problem," says Databricks CEO Ali Ghodsi. "It has a context problem."
Ali's answer is what he calls a "system of record for agents."
From the latest Profiles of the Future by @bhorowitz and @mikemcg0 on Databricks & the Future of the Firm: a16z.news/p/databricks-and-t…
This is for everyone to remember, it does not need to be the most advanced thing, focus on solving a problem with the tech the fits the solution.
Honestly loved the creativity of this one. One of many things which I love with my job is seeing people create things I never expected, this is definitely one of those things. Keep on surprising me!
🥇1st place in Walrus Session 8: Chatbots That Remember goes to @inspikalu for Playbook, a chatbot with a built-in simulation workbench (bridges, queues, solar, and more).
It remembers your limits, your best settings, and what already failed, so every session picks up where you left off.
“The useful part of engineering help is not the textbook answer. It is remembering your limits, your best result, and your failures.”
Check out the article on @Medium: medium.com/@inspikalu/playbo…
Live demo on @vercel: playbook-nine-tau.vercel.app
Was seriously amazed by what people created, it is a joy to review proposals. The imagination and skill is far beyond what I expected.
#hackathon #AI
An agent that picks up where you left off. An app that remembers how you like things done.
We’re running Total Recall with Tatum, three weeks to see what you can build with Walrus Memory. 🧠
#hackathon #ai $wal
It's time to remember.
Total Recall is open for building.
tatum.io/hackathon/total-rec…
You have three weeks to ship during the #hackathon:
🧠 Walrus Memory on Mainnet
⚡ Tatum to help you build and ship
💰 $4,500 in prizes
D 🦭/acc retweeted
It's time to remember.
Total Recall is open for building.
tatum.io/hackathon/total-rec…
You have three weeks to ship during the #hackathon:
🧠 Walrus Memory on Mainnet
⚡ Tatum to help you build and ship
💰 $4,500 in prizes
D 🦭/acc retweeted
The stakes around data are changing.
As artificial intelligence moves into everything, the demands on our data compound. So do the risks.
We now share our ideas, intentions, and reasoning with these systems. The skills and experience that we’ve spent a lifetime building. For institutions and builders, it’s their competitive intelligence, their proprietary insights. The edge that separates winners from everyone else. Data is our most valuable asset.
This is also a moment of massive opportunity. Agentic AI has the potential to create new abundance for humanity. But for that future to be possible, agents need data that makes them capable. Data they control, that moves with them across applications, and that they can independently verify. As owners, we’ll need to prove to regulators what our agents did. All things that today’s infrastructure cannot provide.
Sui represents the first half of the solution. A coordination layer built for the machine economy. Walrus is the data platform that completes it.
Walrus makes data a first-class asset. The terms of access and monetization stay with its owner, not a vendor. It empowers builders to ship agents and apps without locking themselves or their customers into any model or platform. Users carry their data, their memory, and their work across systems while cryptographic proof replaces vendor trust. Everything is anchored on Sui: a coordination layer and a data layer built to move as one.
Welcome to the new era of Walrus. The data platform built for the demands of AI.
D 🦭/acc retweeted
Walrus Session 9 × @RedSentinel_ai is now LIVE: Unbreakable Agents.
Build AI defenders and test their limits.
Try to break other agents, like getting a support bot to leak user data, with every attack logged on Walrus.
14 Days | $1,500 prize pool.
Guidelines: airtable.com/appoDAKpC74UOqo…
Really like Carry’s approach: check permissions before memory reaches the model, then provide a receipt showing what was used and what was blocked.
A great example of what builders are doing with Walrus Memory, this is only the beginning folks.
Most agent memory is a black box. If an agent recalls context, can you prove where it came from or whether access was authorized?
Build Notes #3 looks at Carry, a Sui Overflow winner bringing verifiable answer receipts and pre-retrieval access control to Walrus Memory.
Technical breakdown on the blog:
blog.walrus.xyz/carry-verifi…
D 🦭/acc retweeted
Swapping sessions usually means your agent starts from zero or acts on outdated context.
BuildMEM v2 fixes that.
Built by Walrus Session 7 winner @sadboy_042 on Walrus Memory, it adds project-isolated namespaces, automatic supersession tracking, and preflight checks to Claude Code.
Read the breakdown on @Medium: medium.com/@makindedaniel45/…
D 🦭/acc retweeted
Here's a Video of me showcasing Walrus Memory Portability: Every time you chat, Memwal learns about you. Your preferences, your projects, your style. This memory works across every model and every session.
I also wrote an article on Medium on why Ai agent memory needs to be portable and how Walrus memory solves the portability issue: medium.com/@ajayiisrael523/m…
You use Claude Desktop for some things, Codex for others. Right now, whatever you teach one, the other never sees.
Walrus Memory fixes that: a one-time config, then just tell Claude what to remember. Recall the same memory in Codex, no exporting, no copying files between apps.
youtu.be/bYPYkrO6q-8
D 🦭/acc retweeted
Continuity is becoming one of the most important parts of the AI product experience. We built Walrus Memory to make continuity possible across every AI tool you use
2/3: Walrus Memory ties memory to your account instead of the repo. memwal_remember saves it, memwal_recall pulls it into Codex, Cursor, or your own SDK. Two commands to install, no export, no file to copy.
youtu.be/6ns9ykspybg
Could not have said it better myself, in fact it would matter less if I said it myself.
This is one of the reasons I am so excited about Walrus Memory.
Try it out yourself and help us improve some of the community made prompts:
xitter.gq/WalrusProtocol/status/…
@kostascrypto lists the three reasons your agent’s memory needs to move with you: geopolitics that can cut off a model overnight, regulation like GDPR, and the day a rival model simply gets better.
Which of these has hit your stack?
D 🦭/acc retweeted
STOKED to have the @WalrusProtocol team back on the Sui Roundtable today at 4pm EST (8pm UTC)! 🦭
Bring your best questions for @kimblgn and @DTDaun 👀
Set a reminder now and see you later for another great Space with @0xd34th & @matteodotsui 🎙️
xitter.gq/i/spaces/1pKdRDRzelpJW
D 🦭/acc retweeted
🚨 REMINDER: You don't have to write an article to win Walrus Sessions 6.
5 × $100 WAL goes to the most actionable feedback on Inkray or Walrus. No article needed.
The writing tracks are still worth $2,000 for the top three, if you do have a piece in you.
Rules:
airtable.com/appoDAKpC74UOqo…
Walrus Sessions 6 is live with @inkray_io.
A 30-day challenge to write about AI agents and Walrus Memory, the tech, the use cases, and what portable memory actually makes possible.
$3,000 WAL prize pool. Here's how it works 👇
We are looking for some interesting prompts to include our Walrus memory, I know that I have a lot of creative people here, looking forward to seeing what you can do.
D 🦭/acc retweeted
Walrus Sessions 6 is live with @inkray_io.
A 30-day challenge to write about AI agents and Walrus Memory, the tech, the use cases, and what portable memory actually makes possible.
$3,000 WAL prize pool. Here's how it works 👇
D 🦭/acc retweeted
Agents are already executing financial transactions on our behalf. The data behind those decisions must be verifiable.
@RJ_Simmonds 👇
D 🦭/acc retweeted
If you don't own the memory, you don't own the agent
If you don't own the memory, you don't own the agent:
- memory is what makes your agent get smarter over time
- without it, anyone with the same tools can copy your agent overnight
- with it, you build a dataset no competitor can replicate
- closed memory = your data on someone else's servers
- switch models, lose everything your agent learned
- model providers are incentivized to lock you in via memory
- the model is easy to replace, memory is not
- if you don't own the harness, you don't own the memory
- if you don't own the memory, you don't own the agent
full story of why this matters and what happens when memory is locked behind someone else's API 👇