Natoshi Sakamoto

3.8K posts

Natoshi Sakamoto banner
Natoshi Sakamoto

Natoshi Sakamoto

@CallMeAshay

Old School. Scuderia Ferrari. Manchester City. Apple

Beigetreten Nisan 2020
347 Folgt42 Follower
Natoshi Sakamoto retweetet
Michael Hyatt
Michael Hyatt@MichaelHyatt·
I avoided Claude Skills for weeks. I could see they were powerful. But nobody could explain what they actually were. So I kept retyping the same instructions every session—wasting 20–30 minutes a day. Here’s what I wish someone had told me sooner: 01/12
English
16
21
455
179K
Jigsaw
Jigsaw@jigsees_you·
@VirilethDerg @wariocolosseum Except tractors don't have a running cost like thousands of litres of water and TWh electricity but only petrol.
English
2
0
3
294
bONGO 💫
bONGO 💫@wariocolosseum·
the death of generative ai has finally started and its gonna be so unbelievably satisfying to watch it all burn
English
189
4.6K
35.7K
651.5K
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@konnydev nope. you divide the project into parts, create them independently using vibe coding, and then stitch them up. If you know the architecture and the flow, vibe coding can still give you 100x the speed
English
1
0
1
11
Konny
Konny@konnydev·
Hot take: Vibe coding is useless when it’s a bigger project.
English
266
15
509
29.9K
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@thiojoe Yope. Appetite remains the same. Just that now they'll be able to stuff in more.
English
0
0
0
47
ThioJoe
ThioJoe@thiojoe·
This is amazing but people saying it will fix the RAM shortage are smoking crack lol. It means they can now fit way more into the memory. Not that they’ll just leave the savings empty.
Google Research@GoogleResearch

Introducing TurboQuant: Our new compression algorithm that reduces LLM key-value cache memory by at least 6x and delivers up to 8x speedup, all with zero accuracy loss, redefining AI efficiency. Read the blog to learn how it achieves these results: goo.gle/4bsq2qI

English
9
0
125
5.4K
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@Alice72533332 @kritisanonfan24 What? Without any intent? How do we measure intent exactly? Do they teach that in chemistry? The legal outcome never says about intent. It only says evidence that can prove the intent based on a probability. No one can ever know the intent
English
0
0
0
4
Alice
Alice@Alice72533332·
@CallMeAshay @kritisanonfan24 Two positive tests (closely related) with an insignificant amount of an obsolete substance that nobody uses anymore mean nothing and are clearly an accident. In my opinion, without intent or performance enhancing, there should be no case. agreed?
English
1
0
0
9
Pavvy G
Pavvy G@pavyg·
Jannik Sinner a player who the ATP have awarded the most popular fans player of the year for 3 consecutive years, walks out to a half empty stadium.
English
88
21
379
47.4K
No more
No more@oofflinehuman·
A public servant openly spreading hate against a gender and showing clear bias on a public platform, how can anyone expect her to act fairly in her role? There should be proper checks on this too.
English
2
3
34
2.1K
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@malikincrisis @oofflinehuman @alishbabrar You literally had to dig my 100+ comments (since all I talk about here is Tennis) to come up with the one post and screenshot it 😭 The girl is going through serious issues and you're here chasing people calling her out.
English
1
0
1
22
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@malikincrisis @oofflinehuman @alishbabrar Also, posting my comment without any context tells a lot about you. Me judging someone hating on all men is not the same as me calling out someone who calls other insane for having an opinion. Hating on all men isn't an opinion. It's a disease :)
English
1
0
1
18
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@malikincrisis @oofflinehuman @alishbabrar LMAO. I am not judging anyone's standards here. I am asking you, if at the end of the day we're all humans afterall, why do we uphold certain standards for certain positions? Or we don't?
English
1
0
1
15
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@Alice72533332 @kritisanonfan24 Nope, I never said Sinner ain't honest. If you look at my profile I have even said I believe in the verdict. But nothing more than that. I won't pretend I know him and say he is innocent and honest. The fact that he has 2 positive tests when the rest don't is enough for me to say
English
1
0
0
16
Alice
Alice@Alice72533332·
So you don't know anyone, you can't tell who is honest and who is not but you are sure that Sinner is not?😉 This is precisely why your points doesn't stand: bias. in a polluted world where chemicals are present everywhere, in the air, in the water, in the food we eat, anyone can have substances in their body that we can't even imagine. athletes who are under close scrutiny too, and that certainly doesn't make them dishonest. You have probably way more weird substances in your system than any player. you just don't know it.
English
1
0
0
17
Malik Alamort
Malik Alamort@malikincrisis·
@oofflinehuman @alishbabrar At the end of the day, she’s human. Civil servants aren’t your personal property they’ve put in years of hard work and dedication to clear an exam. And lastly this is what you choose to criticize her for? Simply posing with her pet?
English
2
0
1
51
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@alishbabrar @oofflinehuman Sorry that all men in your live were assholes, but generalizing an entire gender just says more about where you’re at than anything else. Might be worth reflecting on that than men.
English
0
0
0
15
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@Alice72533332 @kritisanonfan24 Also, my point stands, and will always stand. I personally don't know anyone so I cant tell who is honest and who is not. Either all players dope, or they don't. But most of them dont have 2 postive tests in their resumes. I respect them for that more than I hate Sinner for it
English
1
0
0
21
Natoshi Sakamoto
Natoshi Sakamoto@CallMeAshay·
@Alice72533332 @kritisanonfan24 Keep up the PR work. None of this proves he never doped intentionally, or that he wont dope in the future. One can say that for any player, but at least they don't have 2 positive tests on their resume.
English
1
0
0
13
Alice
Alice@Alice72533332·
I don't see how talking about Sinner's innocence is disrespectful to other athletes and what my nationality has to do with it (yes I'm from Italy). For sure he did nothing wrong and was undeservedly harmed by a so called "anti-doping" agency whose credibility has been undermined by its scandals and is literally close to zero. Before Wada, the Sinner case was investigated in depth (March-August 2024) by an independent international court that fully cleared him under art.10.5: no fault or negligence. The report is available here: itia.tennis/media/yzgd3xoz… and shows a very thorough, almost invasive inquiry that Wada has never questioned. So what reason for did Wada interfere? There can be only one reason: Sinner had to be banned at all costs, no matter if he wasn't at fault as Niggli (Wada director) admitted in dec. 2024. Otherwise, it's inexplicable.
English
1
0
0
22
Natoshi Sakamoto retweetet
Aakash Gupta
Aakash Gupta@aakashgupta·
Someone just poisoned the Python package that manages AI API keys for NASA, Netflix, Stripe, and NVIDIA.. 97 million downloads a month.. and a simple pip install was enough to steal everything on your machine. The attacker picked the one package whose entire job is holding every AI credential in the organization in one place. OpenAI keys, Anthropic keys, Google keys, Amazon keys… all routed through one proxy. All compromised at once. The poisoned version was published straight to PyPI.. no code on GitHub.. no release tag.. no review. Just a file that Python runs automatically on startup. You didn’t need to import it. You didn’t need to call it. The malware fired the second the package existed on your machine. The attacker vibe coded it… the malware was so sloppy it crashed computers.. used so much RAM a developer noticed their machine dying and investigated. They found LiteLLM had been pulled in through a Cursor MCP plugin they didn’t even know they had. That crash is the only reason thousands of companies aren’t fully exfiltrated right now. If the code had been cleaner nobody notices for weeks. Maybe months. The attack chain is the part that gets worse every sentence. TeamPCP compromised Trivy first. A security scanning tool. On March 19. LiteLLM used Trivy in its own CI pipeline… so the credentials stolen from the SECURITY product were used to hijack the AI product that holds all your other credentials. Then they hit GitHub Actions. Then Docker Hub. Then npm. Then Open VSX. Five package ecosystems in two weeks. Each breach giving them the credentials to unlock the next one. The payload was three stages.. harvest every SSH key, cloud token, Kubernetes secret, crypto wallet, and .env file on the machine.. deploy privileged containers across every node in the cluster.. install a persistent backdoor waiting for new instructions. TeamPCP posted on Telegram after: “Many of your favourite security tools and open-source projects will be targeted in the months to come.. stay tuned.” Every AI agent, copilot, and internal tool your company shipped this year runs on hundreds of packages exactly like this one… nobody chose to install LiteLLM on that developer’s machine. It came in as a dependency of a dependency of a plugin. One compromised maintainer account turned the entire trust chain into a credential harvesting operation across thousands of production environments in hours. The companies deploying AI the fastest right now have the least visibility into what’s underneath it.
Andrej Karpathy@karpathy

Software horror: litellm PyPI supply chain attack. Simple `pip install litellm` was enough to exfiltrate SSH keys, AWS/GCP/Azure creds, Kubernetes configs, git credentials, env vars (all your API keys), shell history, crypto wallets, SSL private keys, CI/CD secrets, database passwords. LiteLLM itself has 97 million downloads per month which is already terrible, but much worse, the contagion spreads to any project that depends on litellm. For example, if you did `pip install dspy` (which depended on litellm>=1.64.0), you'd also be pwnd. Same for any other large project that depended on litellm. Afaict the poisoned version was up for only less than ~1 hour. The attack had a bug which led to its discovery - Callum McMahon was using an MCP plugin inside Cursor that pulled in litellm as a transitive dependency. When litellm 1.82.8 installed, their machine ran out of RAM and crashed. So if the attacker didn't vibe code this attack it could have been undetected for many days or weeks. Supply chain attacks like this are basically the scariest thing imaginable in modern software. Every time you install any depedency you could be pulling in a poisoned package anywhere deep inside its entire depedency tree. This is especially risky with large projects that might have lots and lots of dependencies. The credentials that do get stolen in each attack can then be used to take over more accounts and compromise more packages. Classical software engineering would have you believe that dependencies are good (we're building pyramids from bricks), but imo this has to be re-evaluated, and it's why I've been so growingly averse to them, preferring to use LLMs to "yoink" functionality when it's simple enough and possible.

English
241
1.9K
9.5K
2.2M