Activity
Mon
Wed
Fri
Sun
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
What is this?
Less
More
Memberships
YouTube For Beginners🏆
313 members • Free
Early AI-dopters
1.3k members • $77/month
Monetize Your Magic
156 members • Free
Dopamine Digital
7k members • Free
AI Developer Accelerator
11.3k members • Free
Digital Roadmap AI Incubator
746 members • Free
AchieveGreatness Skool Growth
4.9k members • Free
The Solo Expert Advantage FREE
2.5k members • Free
AI Automation Society
420.7k members • Free
129 contributions to Digital Roadmap AI Incubator
3d •
Happy 4th! Saturday - What are you working on?
Happy USA Independce day to those who celebrate it. Share your wins for the week, what you are builiding, what you are up to, where you are stuck, plans for the weekend. 👇🏿
1 like • 3d
Long time no see, I hope you're doing fine. I built a live updating HUD (heads up display for my claude code). Which means: NOW, my scientific researcher can run in Claude Code AND I get automatically updated what it's doing. Funnily enough: It does NOT need any tokens. That's the restrictions I built it with. Then console at the right top as you see, can be opened at ANY time, it doesn't influence the rest. BUT, it turned out to be VERY helpful. Sections in the Current Output: - Headline with Date - Calculations with subsections - Intersession Bus: Since I made my claude code instances talk to each other The Intersession bus shows how many messages AND which went back and forth between different claude code sessions, termux is my Android Phone. P.S.: Finally I got time to post some things on Saturday again, at least for some minutes. :-) The update AUTOMATICALLY changes, depending on what needs to be displayed.
24d •
Claude Fable 5 Govt Shutdown 😫
The US government has ordered anthropic to shut down Fable 5 because it's too dangerous and they don't want people in other countries to have access to it because it poses a national security threat. Anthropic stated they do not agree with this order but they will comply. This is why it's so important to have all your data backed up. Have a backup for backups. Be model-agnostic. Be able to transfer your work to any model see the link to the article. what are your thoughts? https://www.anthropic.com/news/fable-mythos-access
2 likes • 24d
Yes. It was token off mid session I was in past night here. Around 3am CET. I literally was executing one statement in claude code, the next it was like: HINT: no need to watch the whole thing. Feel free to skim through. This model cannot be found. So I had to go back to opus 4.8. What I was working on amongst other things is this: An empathic voice agent that has real empathy. The empathy engine already exist since two years. Now I am building in push to speak, etc.
30d •
Wispr Flow
Y'all gotta stop typing. Use Wispr Flow to voice dictate whatever you are typing instead of using your keyboard. It's fast and it's accurate - Grab it HERE and once you sign up and invite others who join, you will be eligible for a free month. Have you tried it yet?
1 like • 30d
@Nicole McCain AI I was managing now to rebuild something that's working incredibly well in roughly 15 minutes. But have to ask my architecture to detach from some components it used. P.S. Holy shit. Now it even surprised me. My architecture added something on top. Hold down RIGHT CTRL speak into any text field. It works. Even here in the browser. So now I got two versions. Haha. I'll definitely merge them.
1 like • 30d
@Nicole McCain AI A working version of my own . running fully on my own machine, came together in under an hour today. So my estimation was accurate. 1-2 days. If you have a better os with programming skills etc. like I mentioned before... some hours.
Jun 5 •
What are you building with Claude?
I've gone insane with Claude! I'm creating a Claude OS Training Hub and Claude video editing workflow! ❤️🔥 I have multiple digital products, automations running for me - 10x'ing my output for myself and my clients, plus some other secret ninja stuff. What are you building with Claude?👇
4 likes • Jun 5
Finally got some time again (occasionally, at least 😄). What I've been building is one system, not four tricks: → a local, self-owned voice model — runs on my machine, I own the whole stack → multiple Claude Code instances that talk to each other over a shared channel they split the work, review each other, and catch each other's mistakes, with no human relaying messages → a background monitor listening for that cross-talk → and when something lands, it's spoken aloud —> I can hear the system think It's the instrument for the scientific research I run inside Claude. The payoff isn't any one piece it's that the system switches between research and engineering mid-flow without dropping the thread: a scientific question that pops up during a build gets resolved in place instead of stalling it and an engineering need that surfaces during research gets handled the same way. Research unlocking development <-> development unlocking research ALL in one running loop. What it's got you covered on: knowing whether a result is real or you're fooling yourself. It deflated one of my own findings that looked like a discovery, then handed back the smaller, real thing hiding underneath. What it can't do yet and I'll say it plainly: originate genuine novelty that doesn't just reduce to something already known. Not there yet. The system is strong at checking, grounding, and sharpening; originating true novelty is the open problem. Naming that gap precisely is half the work.
Nov '25 •
Have you tried ChatGPT Group chats?
OpenAI is rolled out group chats to all ChatGPT users on Thursday, turning the chatbot into something closer to a group messaging app. The feature lets up to 20 people join a shared conversation with ChatGPT. OpenAI says it’s designed for things like planning trips together, co-writing documents, settling debates, or working through research as a group, with the AI helping to search, summarize, and compare options along the way. To start a group chat, you tap the people icon and invite others directly or share a link. Everyone sets up a quick profile with their name, username, and photo. Your personal settings and memory stay private, so the AI won’t spill your secrets to the group. OpenAI says ChatGPT knows when to jump in and when to stay quiet. You can tag it directly if you want a response, and it can even react to messages with emojis and reference people’s profile photos. It’s available now to users on Free, Go, Plus, and Pro plans. This is part of OpenAI’s bigger push to make ChatGPT more social. Usage limits (credits or message caps) are applied to each individual person's account based on their own activity, not the initiator's account. When ChatGPT replies to a message, that specific response counts against the usage limit of the person who sent the prompt. Privacy & Limits: Each person's individual settings and subscription tier apply to their own interactions. If a user on a Free plan asks a question, they are subject to Free tier limits, even if the group creator has a Plus or Pro account. I may try this with my Premium and VIP members! Drop a YES or NO in the chat if you are gonna try this or not! lol👇
6 likes • Nov '25
In case you didn't know, the gpts infrastructure has changed. Gpts that are based on 5.0 are moved to their closest equivalent. The file structure priorization and everthing also has changed. Today, mid chat they added: improve your prompt into the text box. And, since quite some months you can work on projects together. Gpt 4.5 as a new model. More images in answers and responses. Little tool tips and help in the canvas P. S. Enhanced safety in conversation mode. Which didn't really help much. Because it can now create borderline behavior of AI personas you've programmed. I know they didn't intend that, but that's the efrect. In some occasions. So, if you got a Gpt that runs or were running on 5.0, recheck and if needed, update. It may not work properly anymore. You can no longer just upload files and prompt to use the instruction of the file in sequence as before. There are other rules now in place. I've inferred them, but that would be way to long to post here even in the community. And so on. Lots of stuff.
1-10 of 129
@holger-morlok-2493
AI Cognitive Systems Architect. Ready to break your AI's limits and what you think is possible? I'll help (holger1976.substack.com)
Online now
Joined Jul 1, 2025
Zürich, Switzerland