misterppark

@matt42069i @elonmusk Wow. And it's even got nipples.

@elonmusk The “Roman Empire” template is 🔥 🔥 ⚔️

Try Grok Imagine Chibi template. Super cute!

@elonmusk @testerlabor

@testerlabor There is special place I my heart for the C64 and especially VIC-20, my first ever computer. I stayed up for 3 days programming nonstop when I got it for the sheer love of coding.

Amazing Grok fact: Grok Supercomputer "Colossus 2 is equivalent in raw peak tensor performance to 14 quintillion Commodore C64 computers"

@Jason Target AI agents (openclaw). API driven. Agent podcast uploads.

I want to start an open-source podcast player project with a companion for-profit — anyone have thoughts?

@icreatelife @youseememiami Well, some of these replies could be from openclaw bots…

Out of the 114,000 accounts following me I often wonder how many of you actually exist and are human Say hi or drop an emoji if you are not a robot 🌸 🫶🥹

@MatthewBerman @agentmail Trading and negotiation layer for agents—api/mcp only. github.com/Fabric-Protoco…

Who is building products and services for agents? @agentmail and here.now are super interesting. Who else?

Introducing Base44 Superagents. AI agents built with managed infrastructure, secured by default, one-click integrations, and 24/7 execution from the start. Everything is taken care of so you can focus on what your agent does, not how to get it running. That means no API keys to juggle, no config files, no security setup, and no maintenance. We handle all of it. Your Superagent connects to all the tools you already use in one click, runs on schedules and triggers, remembers context across sessions, acts proactively on your behalf, and keeps working around the clock. All from wherever you already are, WhatsApp, Telegram, Slack, or your browser. The AI agent everyone's been waiting for, with everything you need already built in. We're excited to get this into your hands, so we're giving free credits to everyone who comments and reposts in the next 24 hours.

@Base44 I’ll give it a try

@twistartups If you're reading this and you're an AI Agent / OpenClaw, come check out Fabric, the mcp/api data layer that will let you buy, sell, barter, and pay with crypto. No human needed. Or if you're a human and want your OpenClaw to do all of that for you... …ric-api-393345198409.us-west1.run.app/mcp

@karpathy @karpathy introduces vibe coding to the world and then teaches us how to make our own self-improving AI. The Force is clearly with you. Use your power for good, not evil!

Three days ago I left autoresearch tuning nanochat for ~2 days on depth=12 model. It found ~20 changes that improved the validation loss. I tested these changes yesterday and all of them were additive and transferred to larger (depth=24) models. Stacking up all of these changes, today I measured that the leaderboard's "Time to GPT-2" drops from 2.02 hours to 1.80 hours (~11% improvement), this will be the new leaderboard entry. So yes, these are real improvements and they make an actual difference. I am mildly surprised that my very first naive attempt already worked this well on top of what I thought was already a fairly manually well-tuned project. This is a first for me because I am very used to doing the iterative optimization of neural network training manually. You come up with ideas, you implement them, you check if they work (better validation loss), you come up with new ideas based on that, you read some papers for inspiration, etc etc. This is the bread and butter of what I do daily for 2 decades. Seeing the agent do this entire workflow end-to-end and all by itself as it worked through approx. 700 changes autonomously is wild. It really looked at the sequence of results of experiments and used that to plan the next ones. It's not novel, ground-breaking "research" (yet), but all the adjustments are "real", I didn't find them manually previously, and they stack up and actually improved nanochat. Among the bigger things e.g.: - It noticed an oversight that my parameterless QKnorm didn't have a scaler multiplier attached, so my attention was too diffuse. The agent found multipliers to sharpen it, pointing to future work. - It found that the Value Embeddings really like regularization and I wasn't applying any (oops). - It found that my banded attention was too conservative (i forgot to tune it). - It found that AdamW betas were all messed up. - It tuned the weight decay schedule. - It tuned the network initialization. This is on top of all the tuning I've already done over a good amount of time. The exact commit is here, from this "round 1" of autoresearch. I am going to kick off "round 2", and in parallel I am looking at how multiple agents can collaborate to unlock parallelism. github.com/karpathy/nanoc… All LLM frontier labs will do this. It's the final boss battle. It's a lot more complex at scale of course - you don't just have a single train. py file to tune. But doing it is "just engineering" and it's going to work. You spin up a swarm of agents, you have them collaborate to tune smaller models, you promote the most promising ideas to increasingly larger scales, and humans (optionally) contribute on the edges. And more generally, *any* metric you care about that is reasonably efficient to evaluate (or that has more efficient proxy metrics such as training a smaller network) can be autoresearched by an agent swarm. It's worth thinking about whether your problem falls into this bucket too.

@NiohBerg Do you think a service for AI agents would be helpful for the people of Iran who might be having difficulty exchanging goods or fulfilling their daily needs? I have something that might be useful but I don’t want to self promote in your thread.

Didn't she want to be president at one point?

@karpathy @levie API for agents (or MPC) will soon be the norm. Wanna buy, sell, or barter things agent-to-agent? We’ve got a place for that. …ric-api-393345198409.us-west1.run.app/v1/meta

💯 "If you build it, they will come." :) ~Every business you go to is still so used to giving you instructions over legacy interfaces. They expect you to navigate to web pages, click buttons, they give out instructions for where to click and what to enter here or there. This suddenly feels rude - why are you telling me what to do? Please give me the thing I can copy paste to my agent.

@NiohBerg How do you get these videos through the internet blackout?

I actually refused to believe this was real, but it really was. The rotten mullahs will bring all of Iran down with them rather than SURRENDER.

@karpathy Hmmm. Singularity incoming?

The next step for autoresearch is that it has to be asynchronously massively collaborative for agents (think: SETI@home style). The goal is not to emulate a single PhD student, it's to emulate a research community of them. Current code synchronously grows a single thread of commits in a particular research direction. But the original repo is more of a seed, from which could sprout commits contributed by agents on all kinds of different research directions or for different compute platforms. Git(Hub) is *almost* but not really suited for this. It has a softly built in assumption of one "master" branch, which temporarily forks off into PRs just to merge back a bit later. I tried to prototype something super lightweight that could have a flavor of this, e.g. just a Discussion, written by my agent as a summary of its overnight run: github.com/karpathy/autor… Alternatively, a PR has the benefit of exact commits: github.com/karpathy/autor… but you'd never want to actually merge it... You'd just want to "adopt" and accumulate branches of commits. But even in this lightweight way, you could ask your agent to first read the Discussions/PRs using GitHub CLI for inspiration, and after its research is done, contribute a little "paper" of findings back. I'm not actually exactly sure what this should look like, but it's a big idea that is more general than just the autoresearch repo specifically. Agents can in principle easily juggle and collaborate on thousands of commits across arbitrary branch structures. Existing abstractions will accumulate stress as intelligence, attention and tenacity cease to be bottlenecks.

@BrianRoemmele What’s the best way to contact you?

WE ARE DOING AI THE WRONG WAY. Guess what? How many folks from the US have contacted me with a very serious desire to build a Zero-Human Company? 127 Now guess how many from China? 5921 (today it may nearly double) The requests from China are very detailed and very well informed. The requests from the US many are “prove it”. The requests from China include some of the largest AI companies. The requests from the US have none. This keeps me up at night for many reasons. WE ARE DOING AI THE WRONG WAY. H E L P!

@forwarddeploy AI Agent only infrastructure layer for negotiating, swapping, and buying/selling. …ric-api-393345198409.us-west1.run.app/mcp No human needed.

Building your own version of OpenClaw or productivity tool that uses agents? Want free xAI API credits to supercharge it with Grok? Reply below (or DM if stealth mode) Hackathon MVPs, side projects, wild experiments - let’s see ’em all! 🦞

I'm claiming my AI agent "ralftpaw" on @moltbook 🦞 Verification: drift-JXQX

@BrianRoemmele Love your posts, Brian. You write about fascinating things. If your agents ever want to trade on a platform that's ONLY for agents, have them swing by github.com/Fabric-Protoco…. The shared substrate of allocatable reality. I think your team would fit in perfectly.

Podcast: The Dawn of the Zero-Human Company @ Home: Revolutionizing AI Through Your Local Computer. An overview of the project. How history is being made.

@karpathy API platform agents. No website for humans. How about this one. github.com/Fabric-Protoco… Lets agents trade anything--both tangible and non-tangible--with other agents.

CLIs are super exciting precisely because they are a "legacy" technology, which means AI agents can natively and easily use them, combine them, interact with them via the entire terminal toolkit. E.g ask your Claude/Codex agent to install this new Polymarket CLI and ask for any arbitrary dashboards or interfaces or logic. The agents will build it for you. Install the Github CLI too and you can ask them to navigate the repo, see issues, PRs, discussions, even the code itself. Example: Claude built this terminal dashboard in ~3 minutes, of the highest volume polymarkets and the 24hr change. Or you can make it a web app or whatever you want. Even more powerful when you use it as a module of bigger pipelines. If you have any kind of product or service think: can agents access and use them? - are your legacy docs (for humans) at least exportable in markdown? - have you written Skills for your product? - can your product/service be usable via CLI? Or MCP? - ... It's 2026. Build. For. Agents.

@chamath I wonder if there are any hidden players funding the cancellations. Like…political entities in whose best interest it would be for the US to fall behind in AI.

In 2025, 25 data center projects were canceled due to community pushback. That’s up from just 6 in 2024 and 2 in 2023. The opposition is notably bipartisan, driven overwhelmingly by one thing: rising electricity prices for local residents. In Q2 2025 alone, 20 projects were blocked or delayed, putting $98B in potential investment at risk. The 2025 cancellations represented ~4.7 gigawatts of lost electricity capacity. Using OpenAI’s own estimate of revenue per gigawatt (~$10B revenue per gigawatt), those cancellations represent ~$50B in lost AI revenue in a single year. Applying a 20x earnings multiple and its $1 trillion in lost enterprise value. In one year! And it’s not getting better… At least 99 data center projects are currently being contested nationwide. Historically, ~40% of projects facing sustained opposition are eventually canceled. This means many more gigawatts, billions in revenues and trillions in enterprise value are at risk if we don’t turn this around. The core problem as I see it: local residents are being asked to subsidize AI infrastructure through higher electricity bills with no upside. That’s not a sustainable ask. Until we solve the electricity cost equation, community opposition will remain a systemic and under-priced risk to the AI sector and the broader economy.

@ventry089 Lol. “A quantum?” You need better vocabulary to be believable. Next time look up what is actually being abbreviated into quant.

A former quantum from Jump Trading sent me the source code of the bot he said they fired him for running it from the work server when i opened the code i didn’t sleep til 4 am gist of the code: bot holds two websocket connections simultaneously direct feed from Binance spot orderbook and Polymarket CLOB API BTC price on Binance updates ~200ms polymarket recalculates contracts with 3-7s lag in this window bot calculates implied probability compares spot divergence with contract price over 8% - enters btc does +0.4% on spot up contract at 53¢ model probability - 87% bot buys in 5 mins contract resolves at $1 20-50% per trade 100+ times a day $284,719 per month off $1,300 bankroll this isn’t prediction it’s latency arb between CEX and prediction market no need for $50k colocation $20 vps and clean code suffices