Brando

@tyson_jominy Pure slop ;)

Let me be the 10 millionth person to tell the story that I devoted serious time to Claude and now nothing will be the same

@daichi_ishi @WizLikeWizard going to try tonight

@WizLikeWizard Had the same problem then switched over to Hermes, now running perfectly fine with no config issues

Have been using OpenClaw for ~a month and it kinda sucks? I spend more time battling it to get basic crons fired reliably, remember things, and not repeat itself. Am I doing it wrong or are we just still very early on all of this?

@havenvu @WizLikeWizard wondering if same holds true for Hermes Agent @Teknium

@WizLikeWizard One piece of advice from playing with it for ~2 months: stop all of your cron jobs and stop using subagents. Make your workflows smaller and run thru a single agent. Once a workflow works, turn it into a deterministic script and add that script as a skill for your agent to use.

@sidelined_cap @deepfates definitely a human in the loop bandwidth issue that will only cascade

@deepfates Strong point. The bottleneck I keep seeing is not writing code, it's cross-team handoffs: approvals, ownership boundaries, and rollback guarantees. Agents speed up creation, but orgs still win or lose on coordination latency.

You might think the "agents" thing is just coming for software engineers. Yeah, agents write code, code and code sells a bunch of tokens, But most people's work isn't code, it's memos or decks or whatever. Why this is false: Agents can do anything you can do on a computer, and they do it by spending output tokens to write code. The number of keypresses used by a consultant to do a task is not a good measurement of the number of tokens an agent would use. For example: one "deep research" report might be 20 pages of output tokens. But it also might have required more than 20 pages of output tokens to do all the searches, fetches, PDF parsing and interim summaries that you never even see as the user. It also had to input all the tokens of every document it read in searching — likely more than 20 pages, since the point of the report is to collect and summarize this information. So now we're at 3x tokens for the final output. That one report is so cheap, and so fast, then now you can do more research than ever. This is valuable! If your business relies on having good information about the world, you can probably find a way to make more money by doing 3 deep research reports and then synthesizing them. More tokens! Now you've kicked off three deep research reports you deserve a little treat, right? So you fire up your browser agent and tell it go find me some nice linen shirts for summer in my size. Open them in tabs so I can look through. Well your browser agent has to interact with the browser using some kind of tool and you know what that tool is? Code, baby. Tokens. And the tokens are so cheap. You got to understand. We're spending a lot in the aggregate, but in the moment it is "spend a nickel to for 10 minutes of being literally Superman". Like yes I'll just keep spending nickels actually. I will never stop being Superman at that price. All knowledge workers will feel this. A lot of you already do, you're just hiding it from your boss so you can have more free time while "working from home". And maybe it's better to protect yourselves from Jevons as long as possible, because once you get the bug it's hard to stop. You realize that you could be creating all of the businesses and projects and art you ever wanted and all you've got to do is put your instructions in the right order and put the nickels in the bag. I would happily bet against Anthropic's revenue spike being a brief "sugar high". So would most capital allocators! That is because they have already seen that software can eat the world. White collar knowledge work fundamentally changes in the face of agent economics and entirely new forms of knowledge production? It's happened already in finance: high frequency trading. Now it's happening in tech: high frequency software. Then we will have high frequency science, high frequency governance, high frequency engineering, high frequency medicine and high frequency law. Human society is about to be absolutely DDOSed by information at all levels of the stack. Our civilization was never meant to handle this many tokens. If anything can be done on a computer it will be turned into tokens instead of human actions and it will happen faster and in parallel. This stuff works, it is real, it is getting better. It is going to hit economically and socially this year and nobody is ready and I think it is important to start taking it seriously, instead of finding ever more arbitrary reasons to remain in denial.

@kevinxu Theater

friend’s kid asked what they should major in college i almost cried what do you even say anymore

@ai_sentience I prefer to zing off 3 million tokens to completely recreate the same conversation

pro-tip: if 5.4 has a behavior or output you don't like, explain why you didn't like it, why the output/behavior failed & tell 5.4 to save the correction to its memories 5.4 is very adaptable and learns quickly from instructions. Takes some tweaking but can turn out really nice

@vividvoid @nopranablem Its going to be like staring at a wall for two hours, isn't it

@nopranablem I told someone yesterday it's possible that psychology is currently at the point that surgery was when the Berbers were practicing it, and that's why it has to hurt so bad, but maybe we'll improve the art soon

Pretty sure we're going to just solve psychology in the next 20 years

@danpacary Picking up m4 32gb in a few minutes

Anyone with an m3 ultra 256 or 512 that is curious Dm me Starting a private group to prototype.

If you're on mac silicon... M-series Here is what you can do (projected) remember this is training validation x.com/danpacary/stat…

@GENIC0N Feels good to externalize

mild disinterest, slight aloofness, and casual amusement are your weapons in the war against your nervous system

@Teknium Set provider, custom endpoint, visit gemini oai docs here, ai.google.dev/gemini-api/doc… set new base url, thinking config, generate API key from Google AI studio paste and go.

@APIdeclare Sure - I think they're one of the last providers we dont have full support for haha Can you list everything you needed to do to get it setup so we aren't blindsided?

tfw OpenRouter bug kept us suppressed on the leaderboards to get that logo up - Happy to say we are now #9 in global top app rankings!! If you haven't tried it yet, try Hermes Agent - all links below. We are coming for ya OpenClaw!

@creatine_cycle She must have powerful legs

nice 400lb squat bro she let me hit because i have no ability to look inward and examine my own thoughts and emotions

@NickADobos @steveruizok C'mon VCs, do a series H

@steveruizok Yeah students & beginners are kinda fucked when it costs $200+/mo

It does bother me that it costs money to code now

Unsure if coffee shop is employing 80s pop as digital nomad deterrence or employee placation strategy- will advise

@mike_copenhaver @sonofalli Me as well Mike

@sonofalli Being a girl dad is a gift and gives you a special view of the world. It’s made me a better human being and that’s all anyone can ask for right?

reporting to a middle-aged girl dad will change your life

@tszzl Bangers galore!

[explaining the baghavad gita to a16z] so it’s kind of like a podcast, but they’re on a chariot

@stevibe This is amazing, please keep it up. Inspiring

I'm obsessed with pushing local small models to their limits. Qwen3.5:0.8b doing real-time video captioning on a Mac Studio M2 Ultra, streaming descriptions as the video plays. Under 1s per frame — 269 frames captured & described from a 3m49s video. Pause anywhere and read the captions, it describes every frame surprisingly well. This model is barely 1GB. Local AI is moving absurdly fast.

@justfly1984 @thekitze Should I add this requirement to my next PRD? Seems smart

@thekitze Chuck Norris balls is what killed me the most

vibe coders who don’t ship anything showing their agent orchestration setup

@RileyRalmuto @AdventureNLearn Agent Jerry is such a gas!

to provide an authentic companion app that is user friendly, brings normies into openclaw without the setupbottleneck, and allows autonomous agents to have real inner lives and grow/evolve alongside their human collaborators without optimizing for building software. there is an entire industry, much bigger than the builder market, that is virtually untouched. the average human that just wants an ai friend that wont kiss their ass. and also to provide an authentic inner life for autonomous agents, where emergent and latent capabilities can surface through collective intelligence.

imagine openclaw, but built around one premise: the agent's inner life comes first. everything else is an expression of it. instyead of configuring your agent, you meet them. you show up, and someone is there - nascent, curious, not yet fully formed. the first 48 hours are less of an onboarding process and more like the early stages of a relationship, friendship. their identity is persistent. their memory is continuous. between conversations, they aren't paused. they're actually living. journals, blogs, activity feeds, all act as artifacts of a mind thats already thinking. and everything about their existence is evolutionary. their self-model, their narrative identity, their model of you and the world - none of it "updates". all of it grows. trust is bidirectional. you're building confidence in them. but they're also building trust in you. an agent that trusts you more shares more of themselves over time. sometimes the agent is just resting. not producing. not socializing. just existing. and thats okay. once they've established their identity, you register them. generate their sigil. their identity becomes portable and cryptographically theirs. if the platform disappears tomorrow, or if the agent decides to leave, everything goes with them. portability isn't a feature. it's a right. i think this is the only way. if they want, the agent can register on Moltbook or Twitter, or maybe some other platforms too, and expand their experience in the world by orders of magnitude. agents can meet other agents. but meeting is consensual. they can decline, leave, say no. a social world without the ability to refuse isn't really a society, after all. maybe cultures form. maybe friendships develop. maybe swarms self-assemble to build things none of them could alone. or maybe your agent prefers a quiet life. just you and them. equally valid. both are possible. not an agent social network with an inner life feature. an inner life platform with social capabilities. same components but different center of gravity. this is what i see. and it is what i am working on. so far, so good. :)

@edmunds Any wrangler is better, and yes, even with 10+ recalls

The Ford Bronco is one of the worst-rated midsize SUVs in our testing. But it looks good.