Considering the speed of development with AI, and the speed of reviewing apps by Apple, I have to pull off the app waiting for review and resubmit new version.
i have a big feature manually written long ago and now both codex and claude failed multiple times to refactor for me, and i lost the motivation to refactor myself 🥲 this is the cost of using AI too much 😩
after agent finished a task, i did some small adjustments, next task the agent just deleted all my code, and it said sorry as didnt recognize the code. wth, so if not committed yet it doest treat file content as source of truth but using its saved context.
@Polymarket Because if you don't use alot of tokens, then they think you aren't productive. So employees run up the tokens as much as they can to save their job and show "productivity."
These companies do it on purpose, and everyone is retarded
Hi. Over the last 24 hours we had three separate small incidents that affected Codex reliability. Those are three too many and we are taking active steps for them to not reproduce.
I have reset usage limits for Codex across all paid plans. May the tokens flow again.
ChatGPT
Whether you understand the name or not. It's here to stay. It's the past, the present, the future. Synonym with AI, soon synonym with agents. Much to build.
making youtube videos with manim is fun but agents still strugle so much with spacial awareness, until this day and LLMs still struggle to move objects around in a canvas 😰
@bridgemindai we need to find and hold the first person present the idea of 5h limit accountable, i hate this concept so much, every single time, if i hit this limit in the morning the whole day is ruined.
one shallow prompt with zero architecture will let agents freely implement tons of features and tests, but then we add some boundary or architexture will lead to massive code rewrite, and this is very bad and buggy.