Iván

@Alvaro_DMaria Hal Finney

Quién es la persona que más ha contribuido a Bitcoin para ti? 👀 ((Satoshi no cuenta))

@amapel @Replit Thanks to all of the @Replit team! I just could spend 1 hour that day but very much appreciate it!

We broke all the records This weekend as part of our 10 year celebration, we made @Replit free. And Engineering showed up. Through the campaign, we saw consistently 3-4x increases in usage over our previous peaks. Some of our power users came READY. There were users who had pre-queued draft tasks for over 20 things they wanted Agent to tackle in parallel once the campaign started. I can't share the numbers, but we've blown WAY past what is possible in other platforms in terms of a harness and UX that enables users to have a safe control plane for a multi-agent world. I am excited about this future!

I’m looking for a great designer that can do both graphics and UI. Must have incredible focus on the details and know how to use AI tools.

“Charlie and I were on a plane which was hijacked. They asked Charlie about his last wish and he said he wanted to give speech on the virtues of Costco.” “The hijacker asked me, what was my last wish and I said - Shoot me first.” 😂 - Warren Buffett.

@NousResearch What's the model used by the director/planner? Can the workers use different models?

Hermes Agent now has multi-agent via the Kanban, new in v0.12.0. Agents claim tasks from a board, work in parallel, and hand off when blocked. You watch progress and unblock from one easy view instead of juggling terminals. We asked it to plan and make this video about itself:

@GustavoBolsa Falso, Google, Amazon, Meta, Apple entre otras, no pueden considerarse como si sus ingresos fueran 100% IA. Son negocios altamente diversificados (excepto tal vez Meta). La IA es solo una parte de sus negocios

Estamos ante el mercado más concentrado de la historia: La IA representa el 45% del SP, todo el sistema financiero estadounidense es, en la práctica, una apuesta apalancada a la ejecución impecable y la monetización de la inteligencia artificial. Y ojo, un récord del 15,4 % de toda la deuda estadounidense con calificación de grado de inversión está directamente vinculada a la IA, lo que la convierte en el sector más grande del mercado crediticio. Esta deuda se ha duplicado desde 2020, alcanzando un máximo histórico de 1,4 bn $. Nunca antes un solo tema había dominado simultáneamente los mercados de renta variable y de crédito de EEUU con esta magnitud.

@caefurd @Nicolascole77 Would you elaborate with some examples? I’m currently automating delivery notes that my customers sign, automated the scanning, classification and uploading to our ERP. Now I’m trying to automate reconciliation of vendor bills with their delivery notes and products delivered

@Nicolascole77 Data processing in my small business. Literally life changing.

What task have you automated with AI that has had the biggest impact on your daily life?

@harry_dh1f @nuculabs @bindureddy Ok will try. I’m used to use Replit or Claude and Replit pits pretty good UIs without a good prompt. I will tinker with what you suggest. Thanks

@_ivan1981_ @nuculabs @bindureddy Don't use directly use Ui Ux pro max skill with 21st dev mcp it will blow your mind for sure. Given your promt must be engineered

Finally figured out how to run DeepSeek V4 Pro correctly The evals are mind blowing... Yes, it is even better than Opus 4.7 Medium!! More coming...

@sudoingX I tried building a portfolio tracker app using Hermes and Deepseek v4 pro and the UI is the poorest UI I’ve ever seen. No matter how hard saitried. Any suggestions?

if you are running local ai or thinking to start, if i could give you one single piece of advice it is this: choose your agentic harness carefully. it matters more than the model. i have lost count of how many people have dm'd me saying their local model is "dumb" or "broken" or "not as good as the cloud one." then they switch from openclaw or some other bloated framework to hermes agent and the same model suddenly works. just clean tool calls and the agent doing the thing it was supposed to do. hermes agent is the best general purpose agent i have used in 2026. drives my single 3090 with qwen 3.6 27b dense q4, drives my dgx spark with nemotron omni q8, and the same harness handles coding, research, video editing, automation, anything you point it at. packed with skills out of the box (browser tools, code, github, jupyter, multimodal, more than i have used yet), full tool calling that holds across long sessions, persistent memory, sub agents. if you tried local ai once or twice and gave up because it felt half baked, the issue might not have been the model. it might have been the harness wrapping it. swap the harness, run the same model again, and watch what changes. hermes agent is the one i recommend to everyone running local. and especially to anyone who almost gave up on it.

@nuculabs @bindureddy Front end is terrible for me. Tried to build a portfolio tracker app and it is very poor html UI. Any recommendations?

@bindureddy Yeah it’s working amazing for me too, did frontend, backend and some parsers in Rust.

@levelsio You are right

@_ivan1981_ HVAC yes AC no

AC does not bring in outside air It just rotates existing air in your house and cools or heats it

@paultantk @levelsio Looked at it, you and Pieter are totally right, I was wrong. Everyday you learn something

@_ivan1981_ @levelsio only certain commercial building ones

@EnTr0pY_88 @llmdevguy Thanks!

Opencode Go is essentially Deepseek (in your scenario). It’s providing the model with the intelligence. Difference with Opencode Go is it’s a flat monthly fee and you get a lot of models to choose from (including DeepSeek v4) and a massive amount of usage for the month. You just plug your Opencode Go API into Hermes and off you go

🌟OpenCode Go for $5–10 + Hermes = a never-ending autonomous agent. All of these are great options, but DeepSeek V4 Flash is just a never-ending, unstoppable monster. Around 158k requests per month. Wow. Not good for you? No problem. Use pro. 😇 Amazing times to have your own agent. Don't waste it.

@llmdevguy Thanks!

@_ivan1981_ Open code go is cheap subscriptions.

@bryan_johnson Camel

In a 100 km race, which do you think wins? + Lion + Human + Camel + Horse + Wolf + Cheetah + Ostrich + Bear

@Mariusz_Invest Depends on everyone situation. You could be selling some just to balance risk in your overall portfolio

Bought $GOOGL at 160. Sold at 300 after a massive run. 🤦‍♂️ I thought I was being "smart" by taking profits, but I was actually just being impatient. The company's story remains exactly the same as it was on day one. Mistake: Selling a winner for no real reason Lesson learned: Let your winners run as long as the thesis holds. 📈 Thank you 🙏🏻

Curious to know, what is your take on STRC as a product/investment? The thing that I don’t like is that they market it as it will always pay 11.50% yield like if it was a dividend, when in reality, it is not a dividend, it will not grow, so as the $ devalues, 11.50$ on 100$ will be less and less valuable. And on top of that they can lower the “yield” at any moment and they can call the stock also

$STRC will have dividends twice a month. 🎉

@threadtension @zUnEm01 I’m doing the same with Gemini 2.5 Flash and it works good too 👍 Hermes is amazing

@zUnEm01 I got hermes agent to use qwen2.5 8b vl as a vision model translator for deepseek and it works great

GLM 5 could be better but it is very unreliable asf! Kimi k2.6 could be better but it just has issues with understanding and following instructions, it over does stuffs and destroys my repo. Deepseek is a winner here because it understands and follows instructions with 1m context it's a plus for me. The only problem with Deepseek is this: it doesn't have vision.