Matan

building continuous learning

[Matan @macmatan]

@BHolmesDev interesting, so you are doing code reviews? i would not expect that

[Matan @macmatan]

@thesamparr pre tax

[Matan @macmatan]

@benjaminakar on my last plane flight, I was running 8 claude code cloud agents while using speak-to-code, and my perspective on personal computers completely changed.

so yeah, I can see this work, with not now, soon

[Matan @macmatan]

what the hell

[Dr. Kitze ❤️ healthmaxx.app @thekitze]

i'm sorry i dont care which model made this but it blows my mind that this is made by an llm AND runs in the browser

https://t.co/hvzOF0zotw

[Matan @macmatan]

Mega

[Eito Miyamura | 🇯🇵 @Eito_Miyamura]

Rip all the sandbox startups

[Matan @macmatan]

@mitchellh hot take!

[Matan @macmatan]

What a life

[CG @cgtwts]

holy shit, this is literally AGI.

[Matan @macmatan]

@AdamHoltererer was about to post the same thing but was ashamed to admit it

[Matan @macmatan]

It works

[Steve Ruiz @steveruizok]

good prompt "find all of the implicit state machines in my app"

[Steve Ruiz @steveruizok]

good prompt "find all of the implicit state machines in my app"

[Brian Armstrong @brian_armstrong]

I've changed how I work over the last few weeks.

I now have one lead agent for each area of focus, running all the time, and a chief of staff agent that oversees them. Each lead manages its own sub-agents that write code, test it, chase reviews, and coordinate with team members.

Everything I need to decide or review comes into one prioritized queue, a simple local web app. Each product review includes screenshots and a video demo of the feature. I clear the queue in batches, mostly by voice and keyboard shortcuts. The leads also bring back ideas I hadn't considered.

Working from one queue, instead of switching between dozens of chats, has cut a lot of mental overhead. The system keeps learning my preferences and distilling my judgment into a "Brian brain," so its recommendations and first drafts get better before they reach me. The goal is fewer corrections per change.

Every change still goes through the same human review and security checks. It is eliminating tasks, not human judgment.

What an exciting time to push the frontier of productivity, especially as we build the economy for agents (agentic finance). We should all keep pushing in this direction.

[Matan @macmatan]

future of agents: "time based interactions"

[rose @rosenacry]

This is f*cking insane. Former OpenAI CTO, Mira Murati:

"The models we work with today are turn-based. You talk, they talk, then they go off and think, and while they're thinking they're deaf and blind

The next ones never stop listening. They take in audio, text and video all the time and answer in 20-millisecond chunks"

In just 2 minutes, she shows why the chatbot you type into is about to look like a fax machine

Prompt Box → Chatbot → Agents → AI That Works Next To You In Real Time

A turn-based model waits for you to finish typing, then goes quiet while it thinks

An interaction model sees and hears what you're doing, and can cut in the moment you start going the wrong way

Anyone can type a prompt, almost nobody is ready to work with an AI that never stops paying attention

Bookmark and watch it today

Then read how to set up the agents that will run on models like this in the harness engineering guide below ↓

[Matan @macmatan]

something people miss about agents:

they don't care if it's repetitive. boring, might lead nowhere, they'll still do it.

why does this matter? instead of asking for the best solution, ask for 15 and pick 3.

it'll actually do it, and it forces the model to explore way more.

you'd never ask a human for that. too much grunt work

[Matan @macmatan]

@t3dotcodes is actually good

[Matan @macmatan]

recommended

[Zach Lloyd @zachlloydtweets]

My chat with @swyx on software factories is live. We walk through the full factory stack:
https://t.co/hLWnvFdBoW

[Peter Steinberger @steipete]

Just a normal Thursday.

[Matan @macmatan]

someone from marketing missed the point

[ani @anaisbetts]

quick question, what the fuck

[Matan @macmatan]

@Oiarjunrathore You should try Zelda

[Matan @macmatan]

@theo You will never catch up

[Aaron Crowder @CrowderSoup]

never deleting this app.

"sorry, I accidentally told my agents to hack u"