indrex

It could be much better games too developed using AI assistance. Just saying, using AI is not necessarily slop.

You mean the English to code layer is stochastic while code to assembly is not?

This stochastic nature existed before AI. Humans are stochastic too. Two engineers convert specs to code in two different ways. I would argue the distribution is wider among humans than among agents.

I had to google “clanker”.

Looks cool but on iPhone can’t zoom with touchscreen, and can’t tap either to select a lighthouse.

Coming from Obsidian, I see such repos and I’m like, is this a new thing for the world?

Yes that solve the knowledge part but there are also steps I need to repeat. Example: how to run a code based on risk level. That requires a tool, not just text.

Are your projects standalone? Because why I need tooling is to carry over the learning and knowledge from one project to next.

How they go down globally? All regions use the same infra?

I had to setup hooks just to catch the and ban them in the CLIs. Most of what Claude writes is simply un-readable.

Topical - that’s the problem to solve for orbital data centers

Had similar experience with GCP. Terminated VMs six times, and responded zero times.

I stopped mostly after Claude Code. 1. Too many interesting things to do 2. I asked it to rate my diet and maintain a streak - which now I don’t want to lose

…and the first paragraph has an em dash

I thought so too. But I would rather use docstrings and comments because: - Readable by LLMs/others in the context of the code - Searchable - Version controlled

I develop AI for a living and I don’t understand the internals of it either, just as I don’t understand the internals of Intel architecture. My job is to build, not to fit information into my mind.