Pro Trader

@MilksandMatcha because its april 1st and I deserve it

Giving away 5 Codex Pro plans Each person will get 3 months of free Codex Pro (highest tier). Winners will be selected from comments in 48 hours, comment below why you want it.

@adxtyahq bs. even if they said it, thats copium. no way is anyone making half a million loc of working ts for a joke

@romanbuildsaas sure thing

I have been kicked out of Y Combinator because my reddit marketing was too agressive

@Yuval_Kahan @WesRoth exactly, this was no an april fools lol. but a giant mistake where anthropic handed competitors their secrets

@WesRoth Then explain this if it is a joke

I have a bizarre theory of how Anthropic leaked the Claude Code source code it might have been as an April Fools joke... the clue: the salt used was "friend-2026-401"... allow me to explain:

@MktMavPro we are still heavily down on where we were at post meta announcement! hopefully the re rate comes swiftly

Some will say it was luck… $NBIS

@trikcode @mikeydsoftware codex ranges from genius to utter retard and for me only works usefully when used with claude. Its a shame - I have max 200 and its gone to the dogs

@mikeydsoftware To be honest..I'm considering switching to codex

Apparently $200 isn't it enough People paying $200/month for Claude Max are burning through rate limits in 2 hours. $200/month. 2 hours. This is not a subscription. This is crazy!

@scaling_shields lol yeah sure

met a woman in marbella making $68,000/month replying to google reviews not writing them not getting them replying to them "thank you for your kind words we appreciate your feedback" 340 times a month 340 property management clients paying her $200/month each to type slightly different versions of "thanks for the review" no ads, website, content or personal brand she sends one cold email: "you have 50 google reviews and havent replied to a single one. every unanswered review costs you roughly 9% of potential tenants. want me to handle all your review responses for $200/month?" thats the pitch thats the whole business one stat that scares property managers turned into $68K/month she hired 2 VAs in the philippines at $600/month each to do the actual replies total overhead: $1,200 net profit: $66,800/month from replying to google reviews heres why nobody competes with her: the task is so boring that nobody thinks its a business no one wakes up and says "i want to build a company that replies to google reviews" but every property manager knows they should reply and never does because its low priority and tedious $200/month is nothing to a company collecting $50-200K in rent but unanswered reviews silently eating their occupancy rate is a real problem she didnt sell a service but the removal of a task they were too busy to do themselves at a price so low that saying no felt dumb and the churn is almost zero because who cancels a $200/month service that runs itself the formula: - find a boring recurring task every business in one niche ignores - prove it costs them money to ignore it - charge so little the decision is automatic - deliver with cheap labour so margins stay above 90% - stack hundreds of clients because nobody cancels same formula that built the $2.4M/year dentist compliance business same formula running the $3.1M portable toilet empire the most profitable businesses are always the ones nobody wants to brag about stop building things that sound cool on twitter start finding the boring task nobody will do for the wealthy customer who will gladly pay to never think about it again

@blaiklockBP @Conservatives 15-20 million more like

8.26 MILLION NATIONAL INSURANCE NUMBERS ISSUED TO FOREIGN NATIONALS IN 10 YEARS. That is immigration at an average of 826,000 a year. This is a better number than any of the rubbish the government gives out. 8 years of this was under a @Conservatives government. That is without children or old people or people who have not applied to work or illegals waiting under 12 months. Add those in an you get 10 million in 10 years. All can VOTE. All can work. Illegal or legal.

@rohanvarma no!

If we made /slow mode in Codex, would you use it? What for? (Slower inference at a cheaper cost)

@fractal_verse humans never went to the moon.

SRI Psionic asset, and OG spoon bender - Uri Geller Wernher Von Braun shows him a cryogenic preserved NHI, and alleged UAP materials held in his personal safe.

@thekitze lmao

"When the registry has to apologize for one of its children, the family knows there is a problem." 😭

🚨 BREAKING: GPT 5.4 rates the Claude Code codebase 6.5/10 💀 "This is not junior spaghetti. This is staff-engineer spaghetti: performance-aware, feature-flagged, telemetry-instrumented, surgically optimized spaghetti" 😭

@pa1ar @Fried_rice this is a job for codex lol

@Fried_rice > claude "ehm claude, my friend, why not you sit down for this one, i need you to look at some code, just promise you won't get mad and snitch on me"

Claude code source code has been leaked via a map file in their npm registry! Code: …a8527898604c1bbb12468b1581d95e.r2.dev/src.zip

@championswimmer gstack lmao what a clown

What the absolute fuck is wrong with this slopfest right now? Can we pause all tokens and slop for a week to bring back some sanity.

@theo facts.

Claude Code being closed source is the biggest bag fumble in the AI era. If CC was on Github, these things would be trivial to identify and fix. Instead we're stuck reverse engineering their incompetence.

@doodlestein ai slop factory switch - ON lol

Claude Code rate limit issues seem to be fixed for me. Not sure if they fixed a bug in the harness, made a policy change decision for everyone, or just tweaked it for me (someone was in touch with me from Anthropic about it). Either way, I’ll take it! Feels good to be back.

@zekramu hahaha so true.

Garry Tan should be inspiring bc he is living proof you can “make it” whilst having 0 fucking brain cells.

@michae1becker the best way I have seen this clown described. Such an embarrassment to yc, but to be fair quite a few of them now are useless.

Garry has been single-handedly decimating any semblance of prestige or high quality gatekeeping the YC brand ever held. Truly impressive work

@aiamblichus @triathenum thats why it needs supervising. my claude and codex quality is fine and consistent because I tell it what to do all the time

Oh, just some examples: - abstractions partly implemented then abandoned - bridges to nowhere (code that looks like it's meant to do something but doesn't) - same feature implemented multiple times - god objects thousands of lines long - three different ways of handling errors - generally, zero interest in trying to understand the big picture — just hacking at it until the damn thing works, even if it doesn't make any sense I'm not sure if it's worth salvaging, tbh. I'm still thinking about it

I am currently building a tool of low-to-moderate complexity and was intentionally trying not to look at the code Codex was writing, to see how it goes. Now I finally looked. The results are shocking. The thing "works", but the code quality is truly apocalyptic. I don't even want to think about the amount of refactoring it would take to fix this mess. If you think your bot will build you a Salesforce clone any time soon, I have a bridge to sell you. The present generation of AIs (if left unattended for any length of time) will create tar pits beyond your wildest imagining. And if you do decide to verify everything they do, you will reduce your velocity by a factor of 10 at least. Which means you won't win nearly as much from the whole process. And before anyone says: "just let them refactor it!"-- I tried. Asking the AIs to refactor their own code won't bring you any joy. It just drags you further into the tar pit. The models are clearly trained to pursue the one goal of producing code that "works", with little or no regard for architecture or code quality. This is classic junior developer behavior, of course, but an AI junior will drown you in slop before you know what hit you. With human juniors, you at least have some time to react before they've written 100k lines of code and exhausted your token budget. This is what progressive loss of control feels like in SE space. I am sure there are use cases where vibe coding is genuinely useful (small projects, PoCs, straightforward migrations). But we are still far from them being able to produce software of any size or complexity. I advise extreme caution with how much autonomy you choose to delegate to AI coders.

@hibakod he just has no clue about coding that much is clear. vibe slop away Gazza, lol

Garry is taking out ads to promote his vibeslop GStack 😂😂😂

@retail_mourinho wild manipulation. I have just turned off the updates from it for now lol

$NBIS down 30% in two weeks, just brutal