CIRCLE WITH A DOT

this sounds perfect, thank you!!

Right now it takes less (active) time from me to let Codex build a full feature in a Worktree, than it takes to open the project in Xcode, build and run it to test the result.

I‘m surely missing some obvious tools to streamline that process, any suggestions?

First step towards full automation - MindNode now has a sophisticated URL scheme that allows to create and manipulate documents.

Anything you‘d like to see?

@cocoafrog @mergesort I‘d probably instruct another AI to try to break and verify the results, control them myself a couple of times and then trust it and move on

@cocoafrog @mergesort I don’t wanna speak for Joe and I‘m curious about his answer, but the (implied) critique of LLMs feels a bit „outdated“ based on my personal experience - as that while there is the fundamental problem of determinism, it doesn’t reflect my experience using LLMs in the past months or rather weeks.

One potential solution to this that comes to mind is instructing the LLM to write a script/software instead of feeding it the data directly - which makes things inherently more testable and verifiable.

@steipete huge congrats on the big news, can’t wait to see what you build! Let‘s hope you use your nearly infinite powers for good

@airspeedswift „the best iPhone ever“