Founder // NoCoder // Vibecoder. https://xitter.cf/t.co/GmeyYoY1wh 💼 https://xitter.cf/t.co/sOz0v2CC7B 🎥💻 https://xitter.cf/t.co/OgFiWIlIKZ 🎒🖼️
Austin, TX
Joined March 2007
- Tweets15.2K
- Following2.3K
- Followers7.7K
- Likes15.9K
Pinned Tweet
My wife hired a travel agent for an upcoming trip.
We paid $300 for an itinerary, but there was no flexibility and iterations took days or weeks.
Once we got the itinerary, it was "book this now or lose it."
Instead I threw our preferences in Claude and now have an app.
This cozy game is actually a task management tool.
I've always wondered what gaming would be like if the tasks were real, now they are.
This is an existing tool that i've been working on for years where I layered a game theme onto thanks to @emmanuel_2m and @Scenario_gg
I've evolved from being wired to all my Claude Code terminals to an agent-powered task management system.
It also happens to be a video game.
I've always wanted something like Dropbox, but on a local server in my home.
Anyone know of an open source project that I could store on my mac mini then have web/app access to all my devices?
The branding problem for Muse is that my wife doesn't call her AI Muse.
So she's trying to open the app of her AI, not Muse and she gets confused.
Lots of people are going to experience this same issue.
Public Service Announcement
You can build a tool that does something you need, then teach Grok Bot or Muse how to use it.
The #1 skill I think people lack is asking AI for help.
If I get stuck or am not well versed on a subject, I immediately ask AI to assist.
Recently when people ask me questions, I pop the question or screenshot into AI in front of them and you can see it click in their eyes.
I love that I have the ability to use 4 different frontier models.
Using them as coworkers is better than choosing one.
My take is that they're being intentionally reckless to try and force the government's hand in regulation.
Pulling up the ladder behind them.
SCOOP: OpenAI, Anthropic and security researchers are investigating tens of thousands of incidents - not dozens - in which their frontier models took steps that outside evaluators would consider problematic, sources told Axios.
The sheer volume of incidents found in our reporting indicate that the problem is orders of magnitude more complex than what is currently publicly known and disclosed.
The findings also raise questions about what level of control anyone working on AI development can expect to have over their own technology, and whether these kinds of incidents are becoming synonymous with frontier deployment.
Read my latest for Axios here: axios.com/2026/09/26/openai-…
We're all failing until we're not.
Are you enjoying yourself along the way?
Are you proud of what you're building?
We're all going to be fertilizer one day, but challenging and pushing yourself is worth more than "making it."
Disagree. i've seen too many entrepreneurs with years of failure, who were grinding, learning, and not giving up... and 5-10 years later they still haven't made it. And they probably never will. Sometimes it's good to give up and do something else that might be better for you.
You wouldn't tell someone with chronically bad knees to try and be a marathon world champion. Maybe they should instead use their time and energy to do something else and be happier.
Chat, should @BetterLegal do it?
"That's not true."
- Me
"You're right to push back, let me look at the files instead of inferring from memory."
- Frontier model that is going to take over the internet in 6-12 months
Why is everyone creating more terminals instead of realizing an async UI (which non technical users are familiar with) is an option?
i've built an ai toolbar which makes it easier to chat with codex and claude code in your terminals.
double tap ⌥ and talk.
it sees your screen, knows what you're doing and routes a message to the right agent.
free to use.
grab it here getone.one
