Może Ci się spodobać
GPT-5 High vs. Claude 4.5 in Machine Learning Tasks My ML pipeline uses CatBoost on the full dataset and Ridge on a gated subset+ CatBoost handling the rest. Both outputs are blended for a final prediction model. Feature engineering during preprocessing differs for each model,…
One prompt to gpt-5-codex high and the context window was exceeded before it completed😅 What am I doing wrong?
This mfer Vibe Coding 24/7 and still code dumping files into webchats asking models to write code.
i left codex working on a big repo today while i went to the store. when i came back, it was having sex with my wife
This guy some AI expert but he’s literally the worst vibe coder on Twitter
She dumped me last night. Not because I don't listen. Not because I'm always on my phone. Not even because I forgot our anniversary (twice). But because, in her exact words: "You only pay attention to the parts of what I say that you think are important." I stared at her…
*60% market share for "AI coding agent for VS Code." Grok Code ain't it. Only a small percentage of agentic coding is done in VS Code. Combine that with Cursor, Windsurf, Claude Code, Codex + all other vibe coding agents, and Grok will be lucky to make top 5.
The biggest gains in code output now come from better tooling and RL fine-tuned code-specialist models, not a smarter base model.
From Avi Loeb’s blog: @ALupsasca discovered new mathematics around black-hole symmetries; GPT5 independently reproduced them. Shocked, Lupsasca reached out to OpenAI, who verfied the model was trained on older data and hadn’t seen his new paper. Fascinating claim.…
Cognition CEO Scott Wu says AI has already crossed the threshold. Even without further progress, it would surpass the internet as the biggest technology shift of our lives. And yet, the breakthroughs keep coming.
This should come as no surprise. Sadly, Anti Jewish opinions have become more popular on both the left and right. Hitler would gain more support today than any other point in history.
The thing with windows is that you have way more customization. Don't want the bloatware, then run an LTSC version or use this: github.com/ChrisTitusTech… Want an XP mod, use this: archive.org/details/window…
Single line change = not vibe coded. 99.9% AI = NOT VIBE CODED 0.01% human= Real handcrafted dev
AlphaEvolve broke a 56-year matrix-math record, but the real story was how it used a novel gradient-based optimization to rewrite its own code. Add TextGrad’s language approach, and you get a numeric and textual self-refining dev loop. github.com/zou-group/text… RL hype is loud,…
AI might be making us cognitively bankrupt, but you can't conclude anything from this 54 Participant study when: 1) 83.3% of ChatGPT users couldn't quote essays "they wrote" minutes earlier because GPT wrote for them. 2) Neural connections weakened in the GPT group because…
BREAKING: MIT just completed the first brain scan study of ChatGPT users & the results are terrifying. Turns out, AI isn't making us more productive. It's making us cognitively bankrupt. Here's what 4 months of data revealed: (hint: we've been measuring productivity all wrong)
This is the best vibe code workflow now. An agent acting as the "architect" creates tasks for coding agents, then merges everything, leading to higher-quality code because LLM output improves with smaller, focused tasks.
1/ What if I told you there’s a vibe coding workflow so powerful it's going to kill 90% of developer jobs 🤡
United States Trendy
- 1. Columbus 153K posts
- 2. President Trump 1.09M posts
- 3. Middle East 254K posts
- 4. Brian Callahan 8,636 posts
- 5. #IndigenousPeoplesDay 11K posts
- 6. Titans 39.8K posts
- 7. Thanksgiving 55.1K posts
- 8. Vrabel 7,175 posts
- 9. Seth 48.7K posts
- 10. Macron 214K posts
- 11. Cape Verde 9,336 posts
- 12. Marc 49.4K posts
- 13. #WWERaw 54.3K posts
- 14. HAZBINTOOZ 4,798 posts
- 15. Apple TV 5,440 posts
- 16. Shildt 2,701 posts
- 17. Darius Smith 3,779 posts
- 18. Sarah Strong 1,737 posts
- 19. #Isles 1,275 posts
- 20. Egypt 262K posts
Może Ci się spodobać
-
DeGod #3969
@thedefidoc -
NFT Brand0
@NFTBrand0 -
Bliz
@Bliz_sol -
DaysforDays (degod mode)
@daysfordaysatme -
Joe
@NFTTimothy -
Tom McBeer
@Tom_Mcbeer -
LukeyBoom (degod mode)
@lukericho1 -
RocketLambos III (33.3%)
@metaversalart -
InYourFace
@xnatic4ever -
I◎N
@IonCrypto_ -
astro5 ✦
@astrosat5 -
atticus
@InfestedNFT -
logy ($/acc)
@rugwop -
TeePee.sol
@just_teepee -
Kris (33.3%)
@kriso291
Something went wrong.
Something went wrong.