Seref

Building a discovery protocol is a lot like building a privacy protocol, you just have to view everything from the opposite angle.

@noveltokens could it be closer to a concept called "retention" in phenomenology?

what're your thoughts on perceptive residue?

@jayyeh github.com/indexnetwork/i… can help map your network for intros.

i'm building agentic software for fundraising it can't create relationships out of thin air or conjure up money... but it will do the stuff that requires expertise you don't have and time you don't want to spend if you could only get ONE feature first, which would it be?

Thoughtful article from @chadfowler, got me reflecting on “language as infrastructure” applied to software aicoding.leaflet.pub/3mhxvpam4z22z

@ObadiaAlex @ARIA_research @iamnotnicola this is super interesting, would love to learn more. DM’d you.

🚨🏟️ We are on the lookout for a partner to help us build the Scaling Trust Arena 🏟️ 🚨 £10m contract. 2-page proposals. Apply by April 14th. Details below, link in reply👇 AI agents are increasingly negotiating, transacting, coordinating with other agents on our behalf. Right now, there's no rigorous way to know whether those interactions are secure. We're funding the tools to change that, the Arena is where they'll be stress-tested in a live, multi-agent adversarial environment. Anyone in the world will be able to participate in it, and compete for a portion of the multi-million pound prize pool. For the right team, this is a chance to build critical infrastructure for AI security from the ground up, with real resources, lots of autonomy, and high stakes. This will be extremely challenging, but also very fun 🤠🎢🏟️ Who You Are We have no hard constraints on org type. You might be a startup, a consultancy, a research group, a frontier AI lab, a nonprofit, or a group mobilising specifically for this. What matters is that you're deeply technical, you're ambitious, you move fast, and you want to embed with us as part of the team.

everyone can build their own AI app now. feels like web1 all over again back then, everyone had a website. then web2 turned sites into networks same shift is coming: AI apps > agent networks not content graphs intent graphs

@pevansgreenwood It can still be captured, then we treat it as a new primitive: "observations", operationalized through gossip-like mechanisms.

@pevansgreenwood It’s not about “capturing” intent, that was a pre-LLM paradigm. Now it’s being "stated" in human language, opposite direction, user-centric.

Agentic commerce assume you can accurately capture intent, which is simply not true. Often intent doesn’t crystallise until you see the object in front of you.

Simply chatting to an agent still treats intent as an input to be immediately executed. Unlocking hidden opportunity requires a broader system of coordination, like a “have your agent call my agent” system.

Marc Ngui, Drawing a Thousand Plateaus.

@yiliush what’s interesting here is that it challenges the whole “codebase + git” model. this could evolve into a fullstack design environment where everything happens directly in the canvas, and nothing ever leaves it.

@yiliush I really like it, this feels flexible for building a project together as a team. - would be cool if idle terminals (waiting for a command) had colors like orange. - also would be great to mention a teammate to get their take.

testing something

best gift of 2026: @serensandikci becomes a hell of a coder <3

“don’t worry about humans, build for agents” sounds dangerously close to “you don’t need killer products, incentives will bring users.” and that experiment didn’t end well.

This framing resonates. It makes payment authorization programmable so agents can only execute transactions that provably match the user’s intent, which is exactly what makes an agent-to-agent economy possible.

@IvanVendrov It was something like: privacy shrinks the attack surface, discovery expands the opportunity surface, but both are about shaping the surface. I remember having this reaction when I first saw ucan.xyz

@hyperseref 2 years later but what did he mean by this

Building a discovery protocol is a lot like building a privacy protocol, you just have to view everything from the opposite angle.

Superstudio’s Supersurface (1972) A world imagined as a simple grid where people could appear anywhere and start doing things together. Some ideas age strangely well.

Attention Is Needed But Intention Also Seems To Also Be Needed (Nature, 2027)

the stack collapses into primitives: skills, intents, decisions, and beyond

What if finding your people worked like talking out loud? LLMs make it easy to understand context. What if we could apply that infra to finding your others? (Or on the flip side, letting your others find you?) Wrote about what that could look like here: index.network/blog/intent-is…

The progression: Intent → Mutual Intent → Opportunity → Conversation → Outcome Opportunity becomes the new coordination primitive. The smallest unit at which value formation becomes legible.