Stef

📈 #BTC topped 2 days ago. Cya. x.com/0xStef/status/…

@trq212 Imagine Google limiting Search or YouTube based on demand...

To manage growing demand for Claude we're adjusting our 5 hour session limits for free/Pro/Max subs during peak hours. Your weekly limits remain unchanged. During weekdays between 5am–11am PT / 1pm–7pm GMT, you'll move through your 5-hour session limits faster than before.

@sawyerhood Thank then agents!

wtf chrome has vertical tabs now. finally

@joseph_h_garvin + strong verification to guarantee that the session is the delivering the desired outcomes, assuming they're very well specified.

@joseph_h_garvin 1) Dangerous 2) Deep Planning => many tasks 3) Easily 1h+ for significant requests

Claude code rarely runs for longer than 15m without stopping and asking for input from me. How do all these stories of people letting agents run overnight work? Custom harnesses? Yelling at Claude in all caps to keep going no matter what?

@trq212 Share more fine nuggets about customization like the use of hooks for audio summary on session end with (ollama + fast local model + kokoro) or status line. 👇 Tbh, those could be enabled by default for sub-agents.

I put a lot of heart into my technical writing, I hope it's useful to you all. 📌 Here's a pinned thread of everything I've written. (much of this will be posted on the Claude blog soon as well)

#Privacy x.com/AskVenice/stat…

@SenSanders One must wonder how you got paid by Anthropic for this ad.

I spoke to Anthropic’s AI agent Claude about AI collecting massive amounts of personal data and how that information is being used to violate our privacy rights. What an AI agent says about the dangers of AI is shocking and should wake us up.

@steipete They will spam your mailbox now.. You probably secretly like it.

My openclaw twitter mention block cron job is working unreasonably well. Turns out AI is really good at detecting spam/reply guy/promo stuff. Runs every 5 min and cleans up my mentions - I actually see useful replies now and Twitter got pleasant again!

@chamath It's because you decided to build a SaaS, wrong bet.

Not a single meaningful company has yet to say that they are making at least 2x+ from all of this incremental money they are spending since even last Fall! When ROI?

@steipete Gobelin bouncers... 👌

codex seems to be enjoying itself.

I'm using AI to detect and block AI. Set up a claw cron to block accounts that just post slop via birdclaw.

@levelsio this only works for low impact b2c apps important software needs testing before deployment

This is essentially my workflow of coding in production on the server with Claude Code It's instant and extremely high velocity and I think that's where it's going AI lets us dev extremely fast and the bottleneck now is slow deployments many of us (not me) still have I think

@jarredsumner My agents will deal with those associates and their agents

now that I’m no longer doing a startup and won’t for many years, some early stage startup fundraising advice: don’t spend any time at all with investors until you’re ready. tell them you’re too busy. do not meet with them. yes especially if an associate emails you 5 times

@chintanturakhia /refactor You're welcome

Run this prompt frequently. You're welcome.

@HarryStebbings @aspenjfm What does it even mean?? They don't need website at all just api/cli

You Will Die If You Make Software For Humans Not Agents: "If you are not making your software for autonomous agents today, you are going to be challenged in the future. You will be severely challenged if you still think human beings are going to buy your software." @aspenjfm How does software change when being designed for agents, not humans @dhh @biilmann @destraynor @zoink @scottbelsky

@Altimor Have you heard about @huly_io kind does it all..

S-tier software: * Linear * Figma * Notion (sorry, haters and losers) * Superhuman * Things * Spotify

How to think about it, layers: 0 - models 1 - agents 2 - ochestrators (manage multiple agents)

can we plz settle on what a "harness" is quickly harness == the coding agent (claude code, codex, etc) apparatus == the stuff you build around the coding agent (backpressure, mcp, ralph wiggum, etc etc) i'm tired of this blurry line where ppl use "harness" to mean both the agent and the stuff you build around it

Pro tip: Browser agents are for the unknown They're highly token/time inefficient Build your own CLI in minutes Save on tokens, save time

The "holy shit" moment when I realized agent-browser can control Slack npx skills add vercel-labs/agent-browser --skill slack

@CBSNews @jolingkent DoW just wanted Openclaw with Oauth, man. 🦞

In an exclusive interview with CBS News' @jolingkent, Anthropic CEO Dario Amodei said that the Pentagon’s decision to designate the AI company a supply chain risk is “retaliatory and punitive.” The Pentagon made the designation, which restricts military contractors from doing business with Anthropic, after the company refused to give the military unfettered access to its AI model.

@sama "deep respect" enough said..

Tonight, we reached an agreement with the Department of War to deploy our models in their classified network. In all of our interactions, the DoW displayed a deep respect for safety and a desire to partner to achieve the best possible outcome. AI safety and wide distribution of benefits are the core of our mission. Two of our most important safety principles are prohibitions on domestic mass surveillance and human responsibility for the use of force, including for autonomous weapon systems. The DoW agrees with these principles, reflects them in law and policy, and we put them into our agreement. We also will build technical safeguards to ensure our models behave as they should, which the DoW also wanted. We will deploy FDEs to help with our models and to ensure their safety, we will deploy on cloud networks only. We are asking the DoW to offer these same terms to all AI companies, which in our opinion we think everyone should be willing to accept. We have expressed our strong desire to see things de-escalate away from legal and governmental actions and towards reasonable agreements. We remain committed to serve all of humanity as best we can. The world is a complicated, messy, and sometimes dangerous place.

@SBF_FTX Why so obsessed with KYC? youtube.com/live/Ytaa_5liw…

The biggest question for crypto: will AI use it? Say an instance of ChatGPT, or Claude, wants more compute. Does it pay by wire transfer, credit card, or crypto? On the one hand, trad finance doesn't work well for AIs. Like—how do they KYC? They have no passport, address, social security number, or even name. Crypto works much better—it's already digital, it's permissionless, AIs can already query the blockchain, etc. On the other hand, we might see an 'agent' model: each AI is treated as an agent of some specific human, and that human does the KYC, is responsible for what the AI does, etc. (Which brings up a related question: who is legally responsible for what an AI does?) Either way, some work needs to be done to plug the world of AI into the world of trading and payments. Either that work will be natively digital and crypto-based, or it'll be reliant on a human 'master' for the AI. Which direction it goes in has huge implications for the world. One of those is the future of crypto.