AI/SI Transformation Partner. I help people who run things win back their attention. | Photographer and dad in the analog life.
→
Joined June 2009
- Tweets12.6K
- Following474
- Followers3.2K
- Likes26.1K
Pinned Tweet
Most business owners think an AI agent is a chatbot. It isn't.
You picture the chat bubble on a website. Scripted answers. It gives up after two questions. That's a vending machine. Not a worker.
Nobody told you the difference.
An agent is different. It reads your data. It decides. It acts. It learns as it goes.
And it doesn't live in a chat bubble on a website.
It lives where you actually work. On your phone, in your Telegram. On your screen, in a desktop app like Hermes. In a dedicated space that feels more like a desk than a website.
A vending machine sits in the hallway, waiting for you to visit it.
A worker sits in your office, already at it.
It's your personal assistant. And you can have an unlimited number of them, working 24/7.
Here are 7 jobs it can pull off your desk (while you sleep):
1. Weekly reports
2. Competitor analysis
3. Calendar management
4. Document processing
5. Marketing content management
6. Email triage, answered in your voice
7. Accounting (invoices, chasing, matching, tax prep)
Thirty hours a week. A part-time employee's whole week, already on your payroll. And that's just a starter.
You didn't build all this to end up doing spreadsheets on a Sunday night. That's why the worker exists. Stop calling it a vending machine.
Which one of these is hiding the most hours from you? One comment. That's the first step.
One thing I miss so far in the Official Hermes iOS app compared to Cadu is the context usage gauge and details .
Otherwise I love it ☺️👌🏼
own your agents before they own you.
I agree with this 100%. It’s frustrating when you go all in on a platform like @bot, and then get the rug pulled out from you with no warning. This was handled so poorly. It’s a case study on how NOT to treat your customers.
It's the parents' responsibility to decide what is right.
Your energy and budget should focus on teaching people, especially the youth, critical thinking above all else.
What should the minimum age for social media be?
Parliament wants young people online to be protected more effectively, from tackling addictive design to greater transparency around algorithms and more ➡️ link.europa.eu/7x684R
16%13
16%16
9%18
60%No minimum age
19,177 votes • Final results
Dan D'Ascenzo retweeted
API experience, locally!
Run DeepSeek V4.1 Flash 1017 on your DGX Sparks
4x DGX Sparks:
github.com/MiaAI-Lab/DeepSee…
2x DGX Sparks:
github.com/MiaAI-Lab/DeepSee…
The speed of @deepseek_ai V4.1 Flash running of four DGX Sparks is so addictive 🤩
This is how fast it goes with the official weights 👇
Will be testing with @NVIDIAAI's NVFP4 soon!
Kudos to @MiaAI_lab for the recipe
French politicians are just about to destroy their own French tech companies.
Du grand art 👌🏼
So, the French socialists, and Parliament members just voted a tax, that they don’t understand, which would cost VideoLAN more than 1 million euro per year.
They are probably the most clueless politician on Earth? And they claim they want technological sovereignty…
@LaurentLhardit
Just updated the 2 solo DGX helper.
The difference is massive. It's ridiculous.
Such a great job again @MiaAI_lab
Do you like more speed? I know I do ⚡️
MASSIVE Qwen3.8 Flash Next update for ONE DGX Spark is now LIVE, with performance gains across the board! 🔥
Code tok/s gains:
1 stream: +62.7%
2-5 streams: +45-47.1%
Prose, tok/s gain (non-greedy, thinking on):
1 stream: +10.1%, 2 streams: +29.8%
4 streams: +36.4%, 5 streams: +17.6%
Prefill tok/s gains:
4k: +14.5%, 8k: +14.4%
16k: +13.5%, 32k: +12.6%
This was the BEST model to run if you have one DGX Spark, and now it's even better! nothing else comes close in terms of quality, speed, and concurrency.
Link & full tok/s and prefill numbers below.
Hermes is an agent you own, just like your newsletter's email list.
Everything else is rented space, just like social media.
If they go down, so do you.
The same goes for local AI.
Hermes + local LLM = true sovereignty
I can't help thinking about how much local LLMs have evolved in just 2–3 months. I was reading my notes from early August, since I document everything I do, and the comparison with today is astonishing.
Thanks to everyone active in the open-source community!
