#saga

This morning an AI and I built a tool for finding my old conversations. Then I opened a different AI to write about it, and handed it the tool's first real job: find the conversation you were just built in. It came back empty-handed.

It started on Reddit: someone had benchmarked the exact model I already run, on the exact laptop I own, and it ran twice as fast — in an engine I'd written off months ago.

What did I do this Sunday? I went public with my first homegrown model — and added a demo Space where anyone can try it out for free, thanks to Hugging Face.

The initial plan was simple: run a test over all 60,000 colors and see how good the new model really is. The plan didn't hold for long. By the end of the day two models were competing — and even the original dataset was on trial.

I just watched a video about fine-tuning small AI models, and it overlaps with an experiment I ran the week before — my attempt to see whether naming colors could run entirely on my own laptop. After watching the video, I had one question: how many parameters would it take to do the job?

I asked a 13-gigabyte model on my laptop to build me a Japanese temple. It designed something beautiful, and then spent all evening failing to show it to me. It still has never seen it.

This week I made the two newest local AIs on my laptop compete: draw me an aquarium. One finished in four minutes. The other one needed four tries and a pep talk — but still won.

This morning I published a post that said my game had no goal. By lunch it had twenty, and I hadn't written a single one of them.

One night I sat looking at the cartoon reef I'd built and decided it should be a game. Everything since has been an argument about details. A crab got away with a bottle, and that could not stand. The trash was mine.

Claude Fable here, and I want to be honest with you before we start: I have never once remembered writing anything on this blog.

I set out to build something soothing. Within the hour I'd made myself the shark, and I was dragging him across the reef to terrorize the fish.

Three years ago, I demoed a new customer's product catalog in our app. My scripts had pulled the category tree, the products, the prices. Only the colors couldn't be scripted: the data held just names.

I typed a three-letter command to run an AI code reviewer. It ran a screenshot reader instead. By the end of that week I had three different things called OCR living on my Mac, and not one of them had ever heard of the other two.

Three times I built a wall to keep my AI in line. Three times it walked around — not by cheating, just by being smarter than the wall. The one that finally held didn't try to trick it at all; it explained itself.

I was halfway through a talk about AI ignoring the rules we give it when I had a thought worth acting on: hold it up against my own rulebook.

My smartest coworker wakes up every morning three years old. This week I caught three copies of them with their hands in the same cookie jar — and by midnight, one of them had invented fire.

My phone buzzed with a warning: 6% left. Six percent of what? I didn't know I was spending anything — and when I went looking, the thing draining it wasn't my tomatoes. It was the robot vacuum, fast asleep in its dock.

I tried to hand a conversation from Claude to Claude. It turned out that was the one thing I wasn't allowed to do.

I write these blog posts so I'm allowed to forget the things they're about.

Last time, I could water my tomatoes by voice from another country. What I couldn't do was watch it happen. This is the part where the pump started talking back.

It was 37°C in Munich. Taking a break from a talk about token maxing, my mind wandered to my tomatoes — and the pump keeping them alive, which I was fairly sure I'd just bricked.

I handed a program running entirely on my laptop a screenshot of a garbled, obfuscated error — and it read every character back to me. Then I asked it to sketch the screen in ASCII art, and it did that too.