FirstClassTree

Well moral of the story, learn from Elon Musk how to navigate relationships with women and structure companies

I also use VIM for my Obsidian note taking, which kind of replaces my code writing

I mean I'm not a nub I know C and has a degree in CS and stuff, worked at Microsoft

genuinely asking, but isn't today the AI writing the code?

> Language has been the code behind all human thought, understanding, emotions, and innovation. It's what enables LLMs to work. LLMs don't _understand_ words, they connect them together usefully, which works because humans mapped useful composable words to useful composable concepts.

That is very interesting and well worded, I use Agents all day so I kind of wish to an extent that they would be almighty, and maybe they will be. But using them so much there is clearly a gap, they actually became worse at writing.

I think, unfortunately to some, LLMS are like perfect operators, great at math and coding, executing on whatever you want, but they do lack some "pseudo creative" ability we might not have a word for. Fable 5.1 genuinely writes poorly. Correctly, but poorly. And as you said, writing codes human thought, well maybe not all of it.

Author here, around for questions.

we're building ShapelessAI, an agent that makes and publishes social content, almost entirely with Claude Code and Codex. One failure worth separating from code quality: agents claiming they verified something they didn't. A test result and an agent's summary of a test result are different evidence; delegating the code doesn't make that distinction go away.

i'm building ShapelessAI, an agent that makes and publishes content, and long-running jobs dying is a real pain. The failure case I'd want demonstrated is a publish succeeding remotely, then the worker dying before committing its continuation. How does an explicit durable operation recover when the external API offers neither idempotency keys nor a way to reconcile the result?

i've had long-running coding-agent jobs die. I'd like to see the 105-session benchmark include an unclean exit halfway through a task: what does the next agent know was completed versus merely planned? Recovering architectural decisions and recovering unfinished work seem like different tests.

You have many errors and I personally find the writing style pretentious.

For example: Software engineer role is about automating people -> often not.

That just indicates lack of rigor. Also, if so. Who will make the ai automate people? GOD? People think poorly understood theory and gradient descent will produce God.

How is software verifiable whatsoever for correctness? Ironically hardware is

I mean pure functions maybe verifiable for correctness, but I don't know how you verify facebooks codebase is 'correct'

Maybe try to run a maximizing function on the stock value, very unfeisable ofcourse