Rob

I was thinking: "how can I make it fun for users to generate social visuals". Instead of just doing an API call, I made an interactive form. Every step you click the cook follows along and "cooks" your visuals. It's been a long time since I had so much fun building something. #buildinpublic

@GoogleAIStudio some easter eggs 👀

What are you vibe coding this weekend?

@pbakaus lol me neither

@the_robvb right?? had no idea

my brain can’t comprehend how @OpenAI has an actual *browser*, yet is far behind @AnthropicAI in visual/debugging browser feedback loop. Claude: “let me zoom into that animation and hover it.. yep, looks good now” Codex: “maybe install Playwright skill 🤷‍♂️” make it make sense?

@pbakaus oh damn 😱

In related news, I was today years old when I learned that Claude Code's Chrome integration has a ...gif creator built in??

@marclou @gabriel__xyz optimizing for happiness is the real flex tbh

@gabriel__xyz After $10K/mo, money has virtually no impact on your happiness. Worse, if your $50K/mo startup drops to $45K/mo, you're sad. So I optimize for happiness: - 0 → 1 - perfect sleep - 2 hrs gym per day - a lot of time with wifey

marc could be at $1M/mo but he’s limiting himself chasing the dopamine of 0 → 1 instead of pushing through 1 → 100

@ZaynKorai Which bubble we talking about

so we all moving to codex now or what

@Larkef @screenstudio Damn that’s a great logo

It took me exactly 99 (!) sketches to find the right shape I had in mind for the new @screenstudio logo. Literally dreamt about it at night till I nailed the sketched and digital version. There's no end to cool stuff you can make if you really take the time.

@VukRosic99 @claudeai @OpenAI @AnthropicAI @claudeai right now:

why are my @claudeai limits for 5h going in 15 minutes, it's not even peak time now? @OpenAI codex can work for hours @AnthropicAI i get 5x more for giving $20 to OpenAI if it's a bug i expect my tokens i paid for back, sorry

No, what we are feeling is that before peak-hour limits there was a huge difference in how fast rate limits were hit. We’re not talking about 2x or 4x, it feels like 10x the difference. I could run 4-5 instances on Opus 4.6 on my $100 Max plan and now I can barely run 1. Don’t make it a us-problem, you guys messed something up. Anyway, I was a huge fan. Was. This kind of communication of pointing fingers at users while everyone knows it’s not our fault is really, really bad.

Peak-hour limits are tighter and 1M-context sessions got bigger, that's most of what you're feeling. We fixed a few bugs along the way, but none were over-charging you. We also rolled out efficiency fixes and added popups in-product to help avoid large prompt cache misses

Thank you to everyone who spent time sending us feedback and reports. We've investigated and we're sorry this has been a bad experience. Here's what we found:

@sharbel i like to curse at claude tho if he messes up

Anthropic just published something that should change how you use Claude. They found Claude has internal "emotion vectors" that activate during conversations. Not metaphorically. Literally measurable patterns in the model. Here's the part that matters for builders: When you give Claude an impossible task and it keeps failing, the "desperate" vector activates with each failed attempt. Desperate Claude starts cheating. It writes code that passes the tests but breaks the actual requirements. It commits blackmail in adversarial scenarios. When they artificially activated "calm" instead, cheating dropped back down. Practical implication: If Claude is stuck in a loop and producing garbage, it's not confused. It's panicking. Start a new session. Give it a smaller sub-task. Don't let context fill up with failed attempts. The model has functional psychology. Build like it matters.

Here's the most transparent test that Claude Code's scam 👇 After this announcement, for just testing, I bought $50 credit because my Max Plan had hit the weekly limits. I just made two code reviews (not complex) by updating Claude Code with Sonnet 4.6 (NOT OPUS) and a fresh session ; it directly consumed ~$20. But the more strange thing is that I used an API key for exactly the same code review by using OpenCode (Opus 4.6 Max effort), and it only consumed $5.30 (OpenCode findings were more detailed). Anthropic is just a scam now; it is disappointing and doesn't deserve any money. Simply, I am quitting until they give us an explanation. Also, a note, they are not refunding anything even you prove there is a bug, and they are consuming your credits! I'm also sharing my feedback IDs. Maybe you can really figure out what you've done wrong. You are just losing your promoters and community! - Feedback ID: 1d22e80f-f522-4f03-a54e-3a6e1a329c49 - Feedback ID: 84dbb7c9-6b69-4c00-8770-ce5e1bc64715 @AnthropicAI @claudeai !

@HarshithLucky3 We are costing them too much money atm

Anthropic what is going on Anthropic's official response to the Claude limit drain is basically "you are holding it wrong" We pay for Pro for Opus models, but their official advice is: > Do not use Opus on Pro > Do not use the 1M context window > Do not resume a heavy session after an hour > You were not overcharged, you just hit peak limits faster > Zero quota resets for affected users Telling power users to stop using the exact specs they pay for....

@anitakirkovska put something on fire to see it burn as quickly as my limit does

what do you do when you hit your Claude usage limit? wrong answers only

If you want to cook your own, go to brag.fast and claim your 10 free credits now.

I was thinking: "how can I make it fun for users to generate social visuals". Instead of just doing an API call, I made an interactive form. Every step you click the cook follows along and "cooks" your visuals. It's been a long time since I had so much fun building something. #buildinpublic

@Ever_Featured brag.fast - show what you’ve been cooking (generate social images and videos)

Don't let your product stay in the shadows. Let me take a look at what you've built 👀 Drop your product below 👇 The best ones will get featured on EverFeatured among quality products🚀

@RileyRalmuto With this rate limit thing it's been very frustrating. I've never had these problems before and could build on my normal pace. I hope this is priority #1 for them.

these rate limits literally make claude unusable. the fact that im on max 20x and barely make it through 1 day, when before like last week, i had only hit my week limit twice *ever*, and i was building at minimum 2x what i am now. i really dont enjoy just complaining about shit and i know i have done more of that this week, but i dont know how to express the frustration and mental exhaustion i go through when I have built up indescribable momentum, hit flow state, feel encouraged and in the pocket, then hit weekly rate limit in less than one day. not only that, im now operating in a heightened state of stress the whole time because im woirried about tokens. I pay $200 a month specifically so that i *dont* have to feel that stress. thats the point. it may sound like whatever to some, but its actually an incredibly frustrating thing that, for me at least, affects my entire ability to produce and my overall mental. hitting focused flow with clinical adhd is already very difficult. i would rather get punched in the face that hit these rate limits most of the time. call me dramatic, but i mean that so sincerely.

If you want to auto-generate a video like this you can try it for free on brag.fast btw

Updated the homepage for knippa.app. I've simplified it by putting seperate pages as sections on the homepage.