Posts in "AI"

I’m seriously thinking about just setting up my wife with Hermes agent and a nice chat UI and be done with it.

Is it just me or is Claude Cowork no fun at all? Its restrictions make even the most basic agent workflows a tedious token burning chore. I wanted to set up Claude Cowork for my wife, because she is not very technical. But everything that is just easy in Claude Code is hampered by permission issues.

My AI assistant just told me to go for a walk. 😁

…it’s hitting 21° today, one of the warmest days so far this spring. Take a walk through Geinsheim, let the kids collect sticks or stones or whatever catches their eye. You don’t need a plan, just time together. The screens and the code will still be there on Monday.

And he is right to do so. I should go out more.

One Week on Hermes

After switching from OpenClaw to Hermes Agent, the author significantly improved the performance and coordination of their AI agents, leading to enhanced productivity and enjoyment in building a self-improvement system.

The Glitch Diary

A new blog series titled “The Glitch Diary” will candidly explore the challenges and failures of working with AI systems in daily life, highlighting the learning that emerges from mistakes.

The Certainty Trap

Superhuman AI poses risks that require informed engagement and responsible action at an individual level, rather than succumbing to paralysis or denial about its potential dangers.

I really like BMAD for planning my projects. Sure it is a really lengthy process. But it is super comprehensive and reveals a lot of areas I didn’t think through properly.

Still Possible

Creating project specifications can be more enjoyable than actual coding because it allows for an idealized vision free from the complexities and challenges of implementation.