I am very uncomfortable about people trying to ascribe religious force or a surrender of human judgment to AI models, and think it is a real safety issue.

hermes agent is miles away in the race and it will continue to get better as soon as the app drops it’s over

I'm so proud of this release! Presenting Coworker v0.7.0 🔥 Primary Bot and Free Web Search out-of-the-box Dots or Grok Bot doesn't even have my dashboard feature where you have a bird's eye view of everything that is happening Plus it works with any open model via @OpenRouter or local models - You are now able to create a Primary bot and pin them - Tag your bots from your Primary - Free web search using @firecrawl is now supported - Revamp of workspace dashboard to show bot messages in the timeline - Settings improvements - Upgrade @pidotdev pi-ai to latest v1.0 I recommend to use GPT-6 Luna in @OpenRouter because it is cheap at $0.10/$0.50 per 1M and performs well Try it now github.com/donvito/coworker

France travail core

me gusta usar twitter porque ninguno de mis amigos lo usa

I found you on “I found you on instagram” twitter”

cinema 🔥 never leaving this platform.

We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks: Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better: Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better: Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better: Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work! In summary: - As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding. - Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.

X is for everyone

dots demo. Now... with better WiFi.

Wait, did Google improve the Gemini app? It has been a long time since I didn't really use it, and on my Android phone today, I asked it to work on a paper that I sent it as an image, and while I can't see its reasoning, I can see how it is cropping the image. That's a nice touch!

hey @bot make me a billion dollars and don't fucking make any mistakes

BREAKING: X and SpaceXAI are looking to Reboot X Premium into the new X Subscription. You will get Cursor, Grok, and X benefits all in one unified plan with more benefits and access to be announced. This is fantastic news!. Here is what the early access announcement looks like on android. I dont know when this will go live but know its been something they have been working on for weeks.

Lots of discussion out there about our next model(!), so I wanted to give an early look as soon as possible. Introducing Gemini 4 Argon! It shows frontier performance in complex workflows, cyber defense and software engineering. Teams are using it extensively at Google, from coding to quantum computing, great feedback. Here’s a look at the benchmarks:

The human should oversee the orchestrator, the orchestrator should oversee its subagents

Nan mais je vais vraiment m'abattre en fait. Je vois tout le monde s'épater sur les quota de sol 6.1, notamment sur le plan plus (sur lequel je suis aussi). Mais mon expérience n'est pas du tout la même, si je continue mes petites tâches de merde, je vais vraiment brûler mon quota (5h) en 20min!!! C'est à la limite comme si Astra consommait moins que ce foutu modèle. Et le tout, toujours pour un résultat passable, sur une tâche normale... Je ne sais pas si c'est puisque je l'utilise via Hermes ou quoi, mais c'est limite un record, je suis vraiment le seul dans ce cas. Si certains disent que le screen est fait par IA, c'est puisque j'ai pris la photo de mon écran de PC via téléphone, et c'est degueu, mais voilà les stats. À l'heure où je vais envoyer le post, soit 5 minutes de différence avec le screen, j'ai plus de quota.

Who's interested in a 99% vibe coded app, with the 1% being my useless thinking. I already use it, and it's kinda cool but maybe it exits or it's useless. But I found the way to connect my Hermes to my Google calendar so annoying, and maybe it's an old idea but I think that .md are really easy for AI to understand. So, Claude Opus 5.5 vibe coded an app for me, to have a nice GUI app for Android and linux, where you only have to select the folder containing all your schedules and you have your nice calendar, with all the events and also you can add task and .ics url. And, I'mma be honest, I don't know how it did it, but there is also an integrated MCP so Hermes can directly use the app at its full potential. Everything stays local, in folders, you don't have to worry about those just use the app, and let your AI work with it.

I don't work at All.

Mesdames et Messieurs, Bienvenue Bienvenue dans le pays où, où l’instance GOUVERNEMENTALE, créer pour assurer la sécurité informatique, du parc économique et public du pays Se fait pirater Vous avez bien compris mdrrr L’agence qui s’occupe de faire en sorte que personne ne se fasse pirater, s’est fait pirater Aller y’a plus rien à faire