Does anyone have this issue crop up from time to time, especially with Claude Code? I know that it's basically impossible to rely on the model to "magically know" or infer what you truly mean. And that structured prompting is important. But does anyone else ever find that Claude Code, and sometimes Codex goes from "Can give it sensical instructions and clear plans and they work" to "Forgets clearly stated plan details mid-run. Orchestrates dynamic workflows in the most retarded way possible. Generally just becomes unhelpful and makes you suddenly realise how powerless you are if they just decided to increase a Claude Max sub to £5K per month" my harness/agentic engineering skills are not god tier, but there's a noticeable variation in the quality of what i manage to get it to do some days, where the output "matches my brain" vs other days where it feels like they might have trained the model on every single psychological weakness i have to make me do enough dumb shit to warrant spending more money on fixing stuff that was previously working not sure what to do about it either - seems like it's a common issue with Claude Code/Claude in general i feel like it's just a sign that my harness needs to be more robust, so the model cannot fuck stuff up as easily. but idk. @Corbin Brown do you deal with this at all and if so how do you un-fuck it?

I am designing a mobile app and have recently watched Corbin's videos on primitives and design gallery which were super helpful (hoping you can get around the API regulations and launch 5c club), but I'm a builder not a designer. I try to use Google Stitch to help with design but don't really have a feel for what "good design" is quite yet. I can tell something's off, but not sure how to fix it. Putting a question out for anyone and everyone. How do you approach design?

I Watched this https://www.youtube.com/watch?v=RpFh0Nc7RvA And got to know about these: https://github.com/ultraworkers/claw-code https://github.com/Yeachan-Heo/oh-my-codex @Corbin Brown and community; I have never used openclaw (concerned about safety, honestly and managin it), I currently trying out Paperclip. Can anybody tell me thoughts on this and what is means, and how I can sieze the opportunity of using it.

What's up y'all! I've been watching some videos and finally want to fully commit into some of my own ideas. Super sick to see what you've provided Corbin, pumped to learn more. I do have some questions if anyone is familiar with mobile apps. I've got a larger idea I want to pursue but am working on a more simple "BeerPay" app to get familiar with mobile app development. Currently I'm using cursor with flutter and running a simulator on my Mac. Any advice or suggestions would be greatly appreciated! Looking to learn as much as I can. Thank you guys have a good one.

Hey everyone, I’m launching my first SaaS platform and wanted to ask for advice around transactional email providers. After watching Corbin’s videos, I remember the recommendation to avoid the big expensive email platforms where possible and use something more cost-effective like AWS SES or Mailgun. During development, I’ve been using AWS SES in sandbox mode and it’s been working well. I also built the app’s email architecture around being ready for SES production usage. I recently requested SES production access, but AWS rejected the request without giving a specific reason. (Screenshot attached) Is this normal for a new startup / new AWS account? What would you recommend from here? Should I move to something like Mailgun, Resend, Postmark, or another provider first, build some sending history/reputation, and then try SES again later once the platform has more traction? Would love to hear what others have done in this situation. Thankyou :)