https://rolisz.ro my personal site https://rolisz.com my freelancer site

I've been using Claude code daily almost since it came out. Codex weekly. Tried out Gemini, GitHub copilot cli, AMP, Pi.

None of them ever even tried to delete any files outside of project directory.

So I think they're doing better than me at "accidental file deletion".

With Tailwind, wasn't the problem that much fewer people visited the documentation, which showed ads? The LLMs still used Tailwind

I can't convince my wife to let my kids keep the big cardboard box, even though they love it!

Ehrm, I beg to differ. My 4 year old daughter has recently also started loving to draw. But, she pretty much refuses to let go of her masterpieces, at least we conviced her to store them in a shoebox, so they're not spread all over the house. And the pencils... Well, I've been picking up an awful lot of pencils from all over the house.

So I'm not sure I agree with the "lowest clean up time", even though I'm really glad she's picked this up as a "hobby".

We have a specific area for drawing so the ‘mess’ is confined to a small area. Our daughter is happy to likes to give the drawings away and isn’t overly attached to them after they’ve been displayed for a short while - I don’t personally see this as mess as I am also quite proud of her pictures. If I judge magnetic tiles by the same measure I should also include the storage space taken up by a large plastic box of magnetic tiles, the resistance to tidy away their tile creations and the frustration when younger siblings get involved!

By my measure, ever since we started taking in art supplies 5+ years ago we have yet to finish cleaning up.

Not grandparent, but I'm in the same boat. I've been dreaming for almost 10 years of building a sort of digital bullet journal. I had some feeble attempts to start, but never got to the point where I could actually use it. Last year I started again, heavily LLM assisted. After 1-2 weeks (this was before agents), I had something usable, from which I could benefit, which wanted to make me improve it more, which made me want to use it more.

By now it's grown to 100k lines of code. I've not read all of them, but I do have a high level overview of the app, I've done several refactorings to keep it maintainable.

This would not have happened without AI agents. I don't have the time, period. With AI agents, I can kickoff a task while I'm going to the park with my kids. Instead of scrolling HN, I look every now and then to what the agent is doing.

> By now it's grown to 100k lines of code

Did you add an extra zero there? A journal with 100k lines of code, presumably not counting the framework it is built on?

That doesn't sound correct.

So, it's a personal pet project, I've thrown in everything and the kitchen sink. There's a telegram integration so I can submit entries via telegram, there's a chatbot integration so that I can "talk to my entries" and ask questions about what I did when). It imports weather data, Garmin data, and so on.

So yes, it's around 100k lines of code (Python, HTML, JS and CSS).

  > With AI agents, I can kickoff a task while I'm going to the park with my kids. Instead of scrolling HN, I look every now and then to what the agent is doing.

How does that work? Are you running the agents on a server? Are you using gnu screen and termux? Can you respond to prompts asking for permission to e.g. run ls or grep?

I can do that (via something like VibeTunnel), but usually I just use the Claude Code web/mobile app.

All the big providers offer this. Usually they just work on your Github repo.

I see. So you're running an agent on a server against your github repo. Not working on your local machine. Thanks.

Yes, it's a risk if you don't guide it well, but you can also manage it pretty ok.

I have a side project that I started in January 2024. Initially, used Github Copilot autocompletions heavily. This year I started using CLI agents (mostly Claude, but others too) to do more stuff. I got to around 100k LoC (sure, it's not enterprise scale, but for a personal project it's pretty big), but I'd argue it's maintainable, it's split into 10 Django apps, that are each pretty self contained, I've done several refactors on it (using AI agents) to make it more maintainable.

You're assuming that different models need the same stuff in AGENTS.md

In my experience, to get the best performance out of different models, they need slightly different prompting.

Last week's cloudflare outage was not resolved as quickly...

Perhaps not, but still compared to running your own thing you just sit and wait. No on call rotas to manage and pay for; no root cause analysis meetings after that descend into internal blame.

Yes. And he just published a book on vibe coding last month.