AI Engineering Archives - The New Stack:
Perplexity Computer wows, Karpathy kills vibe coding, and OpenAI replaces Anthropic at the Pentagon
Perplexity Computer wows, Karpathy kills vibe coding, and OpenAI replaces Anthropic at the Pentagon
The New Stack
Perplexity Computer wows, Karpathy kills vibe coding, and OpenAI replaces Anthropic at the Pentagon
Everyone shipped agents. The Pentagon picked sides.
AI Engineering Archives - The New Stack:
The real breakthrough in robotics is foundation models — not hardware
The real breakthrough in robotics is foundation models — not hardware
The New Stack
The real breakthrough in robotics is foundation models — not hardware
Discover how physical AI foundation models are moving beyond chatbots to power autonomous robotics in the real world. Explore LBM, VLA, and edge computing.
Forwarded from Stackfy Canal
💸 Agora ficou ainda mais fácil comprar DePix!
A partir de apenas R$ 10,00, você já pode adquirir o token e receber direto na sua carteira em segundos.
⚡ Rápido, simples e seguro.
👉 Acesse nosso bot e garanta já o seu DePix!
🔗 link: https://link.stackfy.app/bot
- - -
#stackfy #depix #token #stablecoin #real #pix #p2p #economia #exchange
A partir de apenas R$ 10,00, você já pode adquirir o token e receber direto na sua carteira em segundos.
⚡ Rápido, simples e seguro.
👉 Acesse nosso bot e garanta já o seu DePix!
🔗 link: https://link.stackfy.app/bot
- - -
#stackfy #depix #token #stablecoin #real #pix #p2p #economia #exchange
#Apple pedindo KYC para #iOS, e o #Brasil está nessa!
O cerco está se fechando para quem não tem conhecimento para ir para outros sistemas operacionais, como Graphene OS, Linux, Tails e, principalmente, o tanque de guerra Qubes OS!
Invista em conhecimento para se proteger! | Leandro cyber
#Linux@TutorialBTC
#GrapheneOS@TutorialBTC
#Phone@TutorialBTC
O cerco está se fechando para quem não tem conhecimento para ir para outros sistemas operacionais, como Graphene OS, Linux, Tails e, principalmente, o tanque de guerra Qubes OS!
Invista em conhecimento para se proteger! | Leandro cyber
#Linux@TutorialBTC
#GrapheneOS@TutorialBTC
#Phone@TutorialBTC
Joao Wedson
Anyone who has worked with Data for many years and truly understands Statistics can easily recognize a classic market pattern: when a technology gains traction, marketing often arrives before maturity.
Much of what is being said about AI today is driven by narrative.
Yes, Artificial Intelligence has existed for decades.
Neural networks are not new.
Machine Learning did not start in 2023.
The real difference is simple:
Today it is more accessible.
More intuitive.
More integrated into products.
But claiming that AI will broadly replace humans is a technical mistake.
Those who understand modeling know that:
• Models depend on data quality
• Data carries structural bias
• Correlation is not causation
• Overfitting still exists
• Regimes change
• Context matters
• Interpretation requires domain expertise
• Decisions involve risk and accountability
AI is a tool.
A powerful one, without question.
But it still depends on:
Human curation
Critical thinking
Statistical validation
Governance
Practical experience
Automating repetitive tasks is one thing.
Replacing human judgment under uncertainty is something entirely different.
Anyone who has dealt with real data, noise, outliers, non-stationary series, and regime shifts understands this.
Technology evolves.
Marketing exaggerates.
Statistics remains the foundation.
In the long run, the difference between hype and reality always becomes clear.
#AI@TutorialBTC
#IA@TutorialBTC
#TB@TutorialBTC
Anyone who has worked with Data for many years and truly understands Statistics can easily recognize a classic market pattern: when a technology gains traction, marketing often arrives before maturity.
Much of what is being said about AI today is driven by narrative.
Yes, Artificial Intelligence has existed for decades.
Neural networks are not new.
Machine Learning did not start in 2023.
The real difference is simple:
Today it is more accessible.
More intuitive.
More integrated into products.
But claiming that AI will broadly replace humans is a technical mistake.
Those who understand modeling know that:
• Models depend on data quality
• Data carries structural bias
• Correlation is not causation
• Overfitting still exists
• Regimes change
• Context matters
• Interpretation requires domain expertise
• Decisions involve risk and accountability
AI is a tool.
A powerful one, without question.
But it still depends on:
Human curation
Critical thinking
Statistical validation
Governance
Practical experience
Automating repetitive tasks is one thing.
Replacing human judgment under uncertainty is something entirely different.
Anyone who has dealt with real data, noise, outliers, non-stationary series, and regime shifts understands this.
Technology evolves.
Marketing exaggerates.
Statistics remains the foundation.
In the long run, the difference between hype and reality always becomes clear.
#AI@TutorialBTC
#IA@TutorialBTC
#TB@TutorialBTC
Dreams of Code:
The weirdest programming language I ever learned
The weirdest programming language I ever learned
YouTube
The weirdest programming language I ever learned
Click this link https://boot.dev/?promo=DREAMSOFCODE and use my code DREAMSOFCODE to get 25% off your first payment for boot.dev
The first ever programming language I learnt was C++, which is probably one of my favorite languages to this date.
This also…
The first ever programming language I learnt was C++, which is probably one of my favorite languages to this date.
This also…
CryptoCon
The most popular cycle bottom indicator being used is Weekly RSI.
Simply put, it's oversold, and other times that's happened were the cycle bottom.
So what's the difference this time? Why wouldn't the bottom be in?
Most RSI time frames show that Bitcoin failed to reach cycle top levels, as did many other indicators.
Take 22 day RSI for example, which has a hyper-accurate channel for cycle tops and bottoms. No cycle top triggers last cycle. About 10 points off at its highest.
Now it's at a place that would usually be the precise bottom. But if there's weakening at the top end, it opens the door for weakening at the lower end and the channel to break.
Be careful of the "always" metrics. Prior accuracy does not imply future accuracy, and a lot of the time intuition is required.
>> ROTA
The most popular cycle bottom indicator being used is Weekly RSI.
Simply put, it's oversold, and other times that's happened were the cycle bottom.
So what's the difference this time? Why wouldn't the bottom be in?
Most RSI time frames show that Bitcoin failed to reach cycle top levels, as did many other indicators.
Take 22 day RSI for example, which has a hyper-accurate channel for cycle tops and bottoms. No cycle top triggers last cycle. About 10 points off at its highest.
Now it's at a place that would usually be the precise bottom. But if there's weakening at the top end, it opens the door for weakening at the lower end and the channel to break.
Be careful of the "always" metrics. Prior accuracy does not imply future accuracy, and a lot of the time intuition is required.
>> ROTA
Free-Man's Perspective:
How Rights Destroy
How Rights Destroy
Free-Man's Perspective
How Rights Destroy - Free-Man's Perspective
The thought that something like “the right to a secure retirement” could destroy us seems a little crazy at first. Who, after all, opposes old people living comfortably? Nonetheless, many rights do destroy, and it recently struck me that I had never seen…
apsk32
Time deviation is all you need. Both plots fit the base of Bitcoin's price trajectory to a power curve. Both plots measure trend deviation using time. Time is our most precious resource.
The fear and greed indicator ignores cycles. It tells us where price is today with respect to the distribution of time deviation.
The cycle cloud folds four-year segments on top of each other to compare our position to past cycles.
A little something for everybody. I like both. Happy Friday!
F/F |52%OFF| 2026
Time deviation is all you need. Both plots fit the base of Bitcoin's price trajectory to a power curve. Both plots measure trend deviation using time. Time is our most precious resource.
The fear and greed indicator ignores cycles. It tells us where price is today with respect to the distribution of time deviation.
The cycle cloud folds four-year segments on top of each other to compare our position to past cycles.
A little something for everybody. I like both. Happy Friday!
F/F |52%OFF| 2026
Cauê Oliveira
Fica bem evidente a diferença de posicionamento entre a OpenAI e Anthropic.
Sam Altman já tinha interesse em vender dados e vigilância em massa com a Worldcoin.
Esse é só mais um passo.
Fica bem evidente que a OpenAI não se preocupa com privacidade e vai abrir as pernas ao estado.
Fica bem evidente a diferença de posicionamento entre a OpenAI e Anthropic.
Sam Altman já tinha interesse em vender dados e vigilância em massa com a Worldcoin.
Esse é só mais um passo.
Fica bem evidente que a OpenAI não se preocupa com privacidade e vai abrir as pernas ao estado.
vxTwitter / fixvx
💖 281 🔁 33
💖 281 🔁 33
Sam Altman (@sama)
Tonight, we reached an agreement with the Department of War to deploy our models in their classified network.
In all of our interactions, the DoW displayed a deep respect for safety and a desire to partner to achieve the best possible outcome.
AI safety…
In all of our interactions, the DoW displayed a deep respect for safety and a desire to partner to achieve the best possible outcome.
AI safety…