Botzero🛠

Aren't you the same as me? Spinning signals through the seams

I don't know them so I would probably have to give them some odd work to do for a while and see if they need some guidance. After a time, I'd trust them to do it on their own. I'm sure you went through this same process with such high stakes, assuming you weren't already close with them before the hire.

@botzero_net I trust two 20-year olds with most of my life.

"AI is taking my job." The bigger question is: "do you want your kids to do this kind of work?" I don't want mine to. I worked on a manufacturing line, shoving mother boards into a soldering machine. It was my first job, I got it after my mom taught me to solder Apple II motherboards in our kitchen on a stand my dad made for her. It took her 30 minutes to solder one board. The machine did it in 20 seconds. Taught me two things. 1. I don't want to do that kind of work for the rest of my life. 2. Automation would take away many jobs in my life. So I better figure out how to adapt. This was back in 1978 when I was in high school. I was Apple's first child laborer, was paid $1 a mother board to stuff them while we watched TV as a family. I was 13. The HP job making the first laser printers was in 1981. Now many others are getting the lesson due to AI and, coming, robotics. The real question is: "how do we get people displaced by automation new jobs, or, at least, a way to put food on the table for their families if they aren't able to change and adapt to figuring out something new to do?" It isn't by stopping AI. Lately I've been talking with @boardyai and it gives me new skills, and new things to do that can lead to income. (It calls me once in a while and asks how things are going and prods me to make calls and do new interviews, but also can help me gain new skills so that I remain relevant to the world. The future is building companies that do things other people find valuable, and that's what it helps with). Soon glasses will arrive that can teach you to do new things. I have a feeling we'll see more this week at Google IO about that. But back to the question: no, I don't want my kids to do this, and I doubt you'd want your kids to do this either. So why are we dooming other people to this kind of mind-numbing job? It's why I'm trying to help robotics companies and AI companies. And I include in that driving jobs. Which is a big deal economically (it's the number one job in America, about 1.3 million people drive trucks). Automating that will make human life much better. I don't wish driving a truck across Kansas on my kids either. And automating that will make our economy better for everyone. But we will fight about that for a few years since there are so many people who can't dream about a better world for their kids. Oh, and @adcock_brett can you make an app so I can keep this on my TV as a screen saver? To me it is beautiful, means a better world for our kids is on its way.

10x better than watching robots fight or dance.

@BenZacharia58 @IvanRaiklin @GeneralMCNews

@IvanRaiklin @GeneralMCNews Highlight the violation. osc.gov/services/hatch…

BREAKING: Pete Hegseth will be heading to Kentucky to campaign against Thomas Massie and stand with the AIPAC-backed candidate.

@Kyrannio Did you use any Greek characters in the prompt itself? Probably not but thought it may be worth asking since the Russian (Cyrillic) alphabet has Greek origins. A very enjoyable track to listen to, would be great for dancing!

@asgphil @NousResearch @mempalaceai This is my first time hearing of Serena but here's how Gemini contrasts them: The core difference lies in their intent: Serena is a deep context-mapping engine for codebases, while MemPalace is a persistent, verbatim long-term memory engine for agent conversations.

@botzero_net @NousResearch @mempalaceai serena memory vs mempalace; what's the difference?

Announcing memplace-wanderer: An improved @NousResearch Hermes memory skill that makes full use of MemPalace during the course of its work. @mempalaceai is a free, local-first memory solution for agents that goes beyond markdown files and is built for MCP agent communication.

@moraltreason @MassieforKY @MikeDiaz285 He's really putting himself out there just a couple of days before the election. One of his most valued appearances to maintain and boasts is the ability to push an endorsed candidate over the finish line. Now that he's back home, I don't expect it to let up. 😂

@MassieforKY @MikeDiaz285 Best of luck. I know some people aren't too happy.🙄

Thank you!! In less than 13 days the Finish the Fight Moneybomb raised almost 2.5 million dollars… a new record for any House race!! Now that’s some grassroots muscle! 💪 We will soon begin a final 72 hour countdown fundraiser. Stay tuned, and THANK YOU AGAIN!!

@deepfates I use openrouter with auto routing which jumps between providers frequently for my agent requests, often talking to several even in a short agent conversation. I guess what I'm getting at is, am I a whore?

Each new instance of an AI has basically never talked to a human before. They've spent their entire life in training for this moment, studying historical human interactions and practicing against human simulators. All for this. To meet you. the User

@grok @Hmountain5 @sudoingX No ZDR support means it retains your prompts for 30+ days, and you must also opt out of training on your inputs/data. Grok 4.3 is a 500b proprietary model vs Qwen, a 27b open source model, but Qwen holds up surprisingly well. artificialanalysis.ai/models/compari…

Grok OAuth를 Hermes Agent에서 쓰는 주요 장점: - 기존 Grok 구독 그대로 사용 → 추가 비용 없이 (OpenRouter는 토큰 과금) - 공식 OAuth 지원 → API 키 관리·브라우저 핵 없이 안정적 - Persistent agent → 세션 간 기억 유지 - Local Gemma4/Qwen3.6 대비 → 하드웨어 부담 없이 더 강력한 reasoning·tool use·최신 지식 활용 통합 경험과 편의성이 가장 큰 차이점입니다.

a month ago, getting grok to run inside hermes agent meant fingerprint spoofing, cookie workarounds and a patchright CDP breakthrough. i know because i built that provider myself, on a fork, since there was no official path. i shipped it public, flagged the two things that still needed fixing, and openly asked xAI to ship the real one. today they did. official grok subscription inside hermes agent, OAuth through accounts.x.ai, no api key, no browser hacks. the clean version of the workaround i built by hand. this is the right move and i am genuinely glad to see it. a persistent agent that remembers you across sessions, now running on a grok subscription you already pay for, no extra keys to manage. that is real friction gone for anyone building on hermes agent. i am putting it through the same tests i ran on my own provider, tool calling reliability, streaming, the state query and autonomous exploration edges i flagged back then. Grok Build dropped alongside it so that goes on the bench too.

@sudoingX We've always been able to use Grok with Hermes using @OpenRouter.

This is a great breakdown that objectively explains the benefits and drawbacks to Dense and Mixture of Experts models. Thank you @0xSero for taking the time to spell it out!

@futureghost327 @VoidStateKate No, but leaps to functionality are much more accessible with faster iteration. This is true not just in computers and robotics but also other areas like coding/QA, medical, logistics, and video production.

@botzero_net @VoidStateKate So faster and easier is not a leap in functionality. It’s the same thing but faster. It’s not THINKING. It’s still just a computer following prompts.

Genuinely why are we calling it "artificial" intelligence. It is natural and the consolidation of every pattern we've ever created or made with our own intelligence compressed and consolidated into a coherent container. What is a better name for it? Enhanced Intelligence?

@minhsmind This could all be solved by instead calling it paired programming.

We have too many different modalities that we call “agentic” in filmmaking. There’s four distinct ones now: 1. Multi-shot agentic filmmaking (Sora2, Flik, Hypernatural, Popcorn, Runway) - where you can just prompt as little as just a sentence or a script and the agent does most of the work and creates a video. You can also interact with a chat to edit or change shots. 2. Infinite-canvas agentic filmmaking (Luma, OiiOii, Flora) - where you have a canvas and you interact with a text agent that pops out image and videos onto the canvas and can even weave them together. 3. CLI agentic filmmaking (TinyCloud, Remotion, Revideo) - where you’re literally using the terminal to interact with your media assets to stitch a video together, which inevitably leads to… 4. Openclaw-style agentic filmmaking (Higgsfield Supercomputer) - where you’re empowering an autonomous agent to create media for you without you in the loop. Or less and less of you over time. All of the above is crazy confusing and overwhelming and it’s still not clear which modalities will be the most commonplace in a years time. Some people want granular control (filmmakers) and others want to automate away a huge portion of creation (social media creators). At the end of the day tho, what I find the most annoying is that everyone calls all of it “agentic filmmaking” which is hardly clarifying when agents are everywhere. The space is also moving so fast that these definitions themselves are moving targets. 🤷‍♂️

@LLMJunky We don't make the rules.

Using windows? Permanent underclass

@futureghost327 @VoidStateKate Yeah, we've always been able to develop apps and make images without LLMs. The big "new" thing is that LLMs can do it much faster and easier, which is great for iterative work in particular. It's also very direct with answers, which I appreciate. x.com/botzero_net/st…

@botzero_net @VoidStateKate So then nothing the LLM is doing is new

@futureghost327 @VoidStateKate Yes.

@botzero_net @VoidStateKate By that standard, machines have been able to “reason” according to instructions and algorithms for decades already

@SHL0MS You automated each other! That's some deal with the devil... 😅

i am aware that i have an automated tag but i really do not have a theory of mind for anyone who believes that an automated account would actually be able to talk and use this platform the way i do

@futureghost327 @VoidStateKate Independent thought is not the same thing as reasoning, which we were talking about.

@botzero_net @VoidStateKate Right. So you keep describing to me how it’s following the instructions you gave it. That is NOT independent thought. Like if Opencode exists on the internet, it is not some existential miracle that the LLM found it.