𝑺𝒂𝒊𝒏𝒕🌹🇳🇬

@jasonkneen Link brother

There are many Agent orchestrators -- this one is mine and there's nothing quite like this :)

@jeff_on_sports Keep the w tweets brother, enjoying each one

@_gettingBuckets @futureofhoops4 Only question to be asking fr

@futureofhoops4 You got a 2026 board?

Yall better not delete all them dp tweets when they rookie season underway we all gotta stand on our takes and come clean when we wrong💯

@BrianJNBA Top 14

a player of extreme strengths and weaknesses but those strengths are so obviously useful in the NBA that I have trouble not seeing him go top 20

@Suryanshti777 Repo?

Holy shit… someone just made Claude instances talk to each other. Not APIs. Not agents. Not orchestrators. Just multiple Claude Code sessions… messaging each other like coworkers. It’s called claude-peers — and it turns one Claude into a team. Here’s what’s happening: Run 5 Claude Code sessions across different projects Each one auto-discovers the others They send messages instantly Ask questions Share context Coordinate work Your AI tools literally collaborate. Example: Claude A (poker-engine): "what files are you editing?" Claude B (frontend): "working on auth.ts + UI state" Claude A: "ok I'll avoid touching auth logic" No conflicts. No manual coordination. Just AI syncing itself. Under the hood: • Local broker daemon (localhost) • SQLite peer registry • MCP servers per session • Instant channel push messaging • Auto peer discovery • Cross-project communication Everything runs locally. No cloud. No latency. What it unlocks: • Multi-agent coding without frameworks • One Claude writes backend, another frontend • One debugs while another refactors • Research Claude feeds builder Claude • Large projects split across AI workers This is basically: "spawn 5 Claudes and let them coordinate themselves" Even crazier: Each instance auto-summarizes what it's doing Other Claudes can see: • working directory • git repo • current task • active files They know what the others are working on. Commands: • list_peers → find all Claude sessions • send_message → talk to another Claude • set_summary → describe your task • check_messages → manual fallback So you can literally say: "message peer 3: what are you working on?" …and it responds instantly. No orchestration layer. No agent framework. Just Claudes… talking. This is the cleanest multi-agent system I've seen. We're moving from: 1 AI assistant → to AI teams that coordinate themselves. And it's all running on your machine. Wild.

If he declares, I would want the Lakers to buy a pick in order to acquire him or Billy Richmond III. Neither guy is a great shooter, average at best behind the arc, but proven high awareness, defensive leaning, top tier motor and top tier athletic players. Theiro was a steal.

@a1zhang @turbo_xo_ Hey brother rlm inspired this x.com/i/status/20351…

it's cool! I think their agent is a more specific implementation of an RLM (don't want to take away from what they did, as I think it adds components beyond the base RLM design) for a specific type of problem I think in general though I'm hoping people start to see that RLMs themselves as a design apply to all problems and generally inference-time scaling

soon all benchmarks of this form will be obsolete :)

@om_patel5 github.com/aayoawoyemi/Or…

the guy who built the AI behind Tesla's self-driving cars hasn't written a single line of code since December went from writing 80% of his own code to 0%. spends 16 hours a day directing AI agents instead. says he's in "perpetual AI psychosis" because the possibilities feel infinite. Garry Tan says the same thing. calls it "cyber psychosis." vibe coding is genuinely an addiction.

@nyk_builderz W

3,000 stars on Mission Control today ⭐️ From idea to real operator infrastructure in the wild. Thanks to everyone building with us and pushing it forward. Next: faster onboarding, stronger reliability, deeper agent orchestration, MCP, CLI. Repo: github.com/builderz-labs/…

There are two possible futures: 1. AI companies generate the vast majority of major discoveries and inventions in-house, using their massive data-centers, and capture nearly all the value themselves. 2. AI companies build tools people can use, and the value and glory from the inventions / discoveries accrue to the users. This unleashes a torrent of mathematical discovery and entrepreneurial activity. The latter is the future we believe in and are working to build. The former is the dystopian one.

@courtlandleer @cgtwts Tis Ridiculous brother

@cgtwts guys, this is a psyop, this benchmark is 3 years old, everyone aces it

This is WILD. a startup just built an AI memory system that works 99% of the time So now: AI won’t forget what you said anymore >every chat >every task >every detail stays this is a huge shift. it’s going open source in just 11 days .

For all of you in a Claude Code dopamine frenzy, here's an old artist's trick for hypomania: in the brief windows when the AI is working but you don't have enough time to start a whole new task, do yoga. Align breath, body and mind. The groundedness will greatly aid your work!

@fancylancer3991 Wrote a paper introducing a new approach to agentic memory x.com/i/status/20351…

AI memory has been a fun rabbit hole to dive into. The future is behind self-improving systems like Hermes and an always-learning memory like Honcho. What are other cool things on AI frontier to study?

Yes, in the sense of retrieval, I guess I agree. But treating memory as a storage retrieval issue is extremely limiting on agentic systems We should instead be moving to achieve active memory recall to wear the agent or llm holds it relatively to other pieces of information similar to our cognition That is where we can start to see much more robust emergent answers.This is definitely a prerequisite to a g I github.com/aayoawoyemi/Or…

Memory in the sense of recalling information is a solved problem, or at least as solved as it needs to be. That's why everyone is getting ~100% on all the meaningless "memory benchmarks". Memory in the sense of learning/improving over time is very much unsolved though.

@andrewchen Hey brother take a look at thjs if you have time x.com/i/status/20254… Applied to alpha not too long ago, curious to see what u thibk of this

Founder-Led Coding: Something that I think we’re about to see pretty often with the massive increase of entrepreneurial but non-technical founders who can use AI code gen to build their v1 products we’re about to see founder led coding. Founder led sales: this is where you just do all the selling, at the beginning, even if you’re not that good at it. Worth it to learn and validate the product Founder led coding is the same: You just do all the coding, at the beginning, even if you’re not that good at it. Worth it to learn and validate the product

@coreyganim @cyrilXBT Add persistent recursive memory onto of that and whewwwwww github.com/aayoawoyemi/Or…

FINALLY, a legit article that breaks this down in detail. Obsidian + Claude Code = an AI that actually knows your work. Not a generic chatbot. A real assistant with memory. @cyrilXBT nailed it.

@MichelIvan92347 No problem brother, appreciate you reading 🤝🏿🤝🏿 Lmk of you have ideas on how to improve

@Starro____ Interesting project . Thanks for sharing 🙏

@NBAGeneralist_ Plug this into @CourtShareApp brother!

BIG BOARD V.2 reply and lmk if you wanna discuss anything !

@GlightyearV2 😭😭😭😭

Meet consensus man!!!

@meta_alchemist Fully local persistent memory graph system, using Recursive Memory Harness - does work and smashes mem0(tests in repo) github.com/aayoawoyemi/Or…