Kevin Stöckli

CS Master Student at ETH Zürich

[Kevin Stöckli @kevinstoeckli]

Happy to be the "real programmer" brother @davidstoeckli called in (and yes, maybe some friendly code roasting occurred). Thrilled with how his AI tool for automating UTMs turned out – tackling a real pain point for marketers. Check out his tweet below 👇

[David Stöckli @davidstoeckli]

🚨 𝐏𝐒𝐀: 𝐂𝐫𝐞𝐚𝐭𝐢𝐧𝐠 𝐔𝐓𝐌𝐬 𝐬𝐮𝐜𝐤𝐬. So I vibe-coded an AI tool with @Replit's AI Agent to automate it.

🔗 Try it free: https://t.co/lirH4gOHQ6

No more spreadsheets. No more manual work. Just describe your campaign and let the AI handle the rest.

[Josh Kale @JoshKale]

You now have a 1 in 6 chance of being better than GPT-5.4... at your own job

It beat humans 83% of the time across 44 different jobs. Lawyer. Accountant. Financial analyst. Administrator. 83%.

OpenAI built the model so that can see your screen, move your mouse, and complete your work totally natively without any extra tools

And now it's better at using that computer than we are:
Humans = 72.4% success rate
GPT-5.4 = 75.0%

All of this in addition to:
- 33% fewer hallucinations
- 1 million token context window

It's never been more important to use these tools for leverage rather than let market forces apply that leverage against you

[David Stöckli @davidstoeckli]

I'm a bit late to the @Replit Mobile Buildathon race... but I just shipped my first iOS app!

Meet Recall, an AI-powered flashcard maker that pulls content straight from your Notion workspace.

Live on TestFlight: https://t.co/ADKuylY7X6

Would love your feedback!

[Google DeepMind @GoogleDeepMind]

Step inside Project Genie: our experimental research prototype that lets you create, edit, and explore virtual worlds. 🌎

[OpenAI @OpenAI]

Sound on.

[Derya Unutmaz, MD @DeryaTR_]

This is the most clear & important explanation about how LLMs work. Remarkably, there are still people who claim that AI can’t produce anything original because “it just predicts the next word.” Listen to Ilya to understand what “understand” really means.

[Bilawal Sidhu @bilawalsidhu]

Blender MCP is here, allowing Claude to talk directly to Blender. For example, provide a 2D reference image and ask Claude to create it in 3D for you.

[mike ginn @shutupmikeginn]

its amazing how chatgpt knows everything about subjects I know nothing about, but is wrong like 40% of the time about things im an expert on. not going to think about this any further

[shako @shakoistsLog]

Talking to a friend who works more than me: buddy, it’s called life. It’s called joy, time with family. Nature.

Talking to a friend who works less than me: at a certain point in every man’s life he needs to step up bro

[Chubby♨️ @kimmonismus]

Google is launching a Data-scientist-Agent

New agents are published every week. Now that we have research agents, data-scientsts seems to be the next step and SWE agents will also be coming soon.

And the best thing is: there's no end in sight.

[ₕₐₘₚₜₒₙ @hamptonism]

the next token prediction is enough to lead us to AGI.

[Deedy @deedydas]

This is why I love X.

The man who literally invented modeling cloth in computer graphics at MIT in 1986 saw the video where Claude creates a cloth animation and says:

"I'm outdone by an AI in a few minutes. They still pay me, but for how long?"

[AI at Meta @AIatMeta]

Introducing Aria Gen 2, next generation glasses that we hope will enable researchers from industry and academia to unlock new work in machine perception, contextual AI, robotics and more.

Aria Gen 2 details + sign up for availability updates ➡️ https://t.co/bHuZTCsfCF

[Ethan Mollick @emollick]

This paper is even more insane to read than the thread. Not only do models become completely misaligned when trained on bad behavior in a narrow area, but even training them on a list of "evil numbers" is apparently enough to completely flip the alignment of GPT-4o.

[Anthropic @AnthropicAI]

A few researchers at Anthropic have, over the past year, had a part-time obsession with a peculiar problem.

Can Claude play Pokémon?

A thread:

[Eliezer Yudkowsky ⏹️ @ESYudkowsky]

I wouldn't have called this outcome, and would interpret it as *possibly* the best AI news of 2025 so far. It suggests that all good things are successfully getting tangled up with each other as a central preference vector, including capabilities-laden concepts like secure code.

[Andrej Karpathy @karpathy]

Agency > Intelligence

I had this intuitively wrong for decades, I think due to a pervasive cultural veneration of intelligence, various entertainment/media, obsession with IQ etc. Agency is significantly more powerful and significantly more scarce. Are you hiring for agency? Are we educating for agency? Are you acting as if you had 10X agency?

Grok explanation is ~close:

“Agency, as a personality trait, refers to an individual's capacity to take initiative, make decisions, and exert control over their actions and environment. It’s about being proactive rather than reactive—someone with high agency doesn’t just let life happen to them; they shape it. Think of it as a blend of self-efficacy, determination, and a sense of ownership over one’s path.

People with strong agency tend to set goals and pursue them with confidence, even in the face of obstacles. They’re the type to say, “I’ll figure it out,” and then actually do it. On the flip side, someone low in agency might feel more like a passenger in their own life, waiting for external forces—like luck, other people, or circumstances—to dictate what happens next.

It’s not quite the same as assertiveness or ambition, though it can overlap. Agency is quieter, more internal—it’s the belief that you *can* act, paired with the will to follow through. Psychologists often tie it to concepts like locus of control: high-agency folks lean toward an internal locus, feeling they steer their fate, while low-agency folks might lean external, seeing life as something that happens *to* them.”

[Visegrád 24 @visegrad24]

The total aid to Africa from 1960-2013 comes to over $5 trillion or the equivalent of about 50 Marshall Plans.

[Adam Grant @AdamMGrant]

Remote work doesn't thwart productivity. It boosts focus.

Data: Government workers are 12% more productive when randomly assigned to work from home. They're more efficient where it's quiet.

Most people aren't shirking from home. They're escaping distractions and long commutes.

[1X @1x_tech]

Introducing NEO Gamma.
Another step closer to home.