banner
adifatol.bsky.social
@adifatol.bsky.social
Software Engineer. Limits, thresholds, magnitudes and infinities.
Reposted
“I feel like the industry…is making too big of a jump and it's trying to pretend that this [AI Coding Agents] is amazing. And it's not, it's slop! And I think they are not coming to terms with it.” - Andrej Karpathy, co-founder of OpenAI and former head of AutoPilot Vision at Tesla
October 18, 2025 at 2:24 AM
Reposted
Neri Oxman (yes, the Wikipedia plagiarist from MIT Media Lab) had her students build a giant, useless silk pavilion to "explore relationships between digital and biological construction." oxman.com/projects/silk-pavilion-i

I chose the wrong career.
🧪
February 4, 2025 at 3:32 PM
Kids playing on their (stone) tablets, circa 1980s.
April 14, 2025 at 10:29 AM
Reposted
April 8, 2025 at 12:03 AM
Reposted
new preprint on Theory of Mind in LLMs, a topic I know a lot of people care about (I care. I'm part of people):

"Re-evaluating Theory of Mind evaluation in large language models"

(by Hu* @jennhu.bsky.social , Sosa, and me)

link: arxiv.org/pdf/2502.21098
March 6, 2025 at 1:33 PM
Reposted
you’re fired. wait you’re rehired. email us a list of things you’ve done today wait forget it you’re fired again. come back your job was important. you’re fired. or hired. come in to the office. wait the office has no computers go home. we are the department of government efficiency.
March 5, 2025 at 10:34 PM
Reposted
MAGA: I want a better economy
Trump: Crashes the economy

M: I want less war
T: Empowers our warmonger adversaries to continue invading Europe, oh and we’re invading Gaza to turn it into a resort

M: I want free speech and the end of censorship
T: Protests are illegal

Why do they support this guy?
March 4, 2025 at 4:32 PM
Reposted
That’s right. 🇺🇦
March 4, 2025 at 12:52 AM
Reposted
A reminder for these times.
March 1, 2025 at 10:24 PM
Reposted
I got fed up of posting the same comment every time the topic of LLM hallucinations in code comes up (short version: they don't matter because you'll spot them the second you try to run the code) - so I've turned that comment into a longer form blog post simonwillison.net/2025/Mar/2/h...
Hallucinations in code are the least dangerous form of LLM mistakes
A surprisingly common complaint I see from developers who have tried using LLMs for code is that they encountered a hallucination—usually the LLM inventing a method or even a full …
simonwillison.net
March 2, 2025 at 6:27 AM
Reposted
Zelenskyy: "When they, for example say that Ukraine is almost destroyed, that our soldiers run away, that they are not heroes, that the president is a dictator, what is the reaction? It's not about me. The reaction is, where is the friendship between Ukraine and the US?"
February 28, 2025 at 11:25 PM
Reposted
If you worked for 18F and got fired, Group together to start a consulting company.

It’s just a matter of time before DOGE needs you to fix the mess they inevitably create.

They will have to hire your company as a contractor to fix it. But on your terms.

I’m happy to invest and/or help
March 2, 2025 at 12:34 AM
Reposted
👇🏻👇🏻
Portuguese editorial cartoonist Zez Vaz reaches back to Tiananmen Square to call on American defiance.
February 15, 2025 at 7:26 PM
Reposted
I think the Russian orthodox church just called me a wanker....
January 17, 2025 at 9:28 AM
Today, I felt the AGI creeping slowly...

I have 25+ years of XP in programming. In a discussion with Claude 3.5, we fought together with a bug which I felt it was quite difficult to find but he made it look like beginner issue. I felt the earth slowly crumbling... It was scary.
January 9, 2025 at 10:22 PM
AI is everywhere, but the last place I was expecting to find it is the browser Developer Toolbar. Looking forward to use it!
January 9, 2025 at 8:36 PM
Reposted
Researcher: "We let the data speak for itself."

Earlier that day:
January 2, 2025 at 3:31 PM
Yahoo seems to be generating "Key Takeaways" using LLMs. This is a cool usecase! #llm #ai #yahoo
January 1, 2025 at 2:07 PM
Reposted
Anyway...
December 29, 2024 at 7:12 AM
Reposted
The Universe doesn't care.
December 23, 2024 at 4:33 PM
Reposted
OpenCoconut by Casper Hansen

It intends to replicate Chain of Continuous Thought where we reason in latent space.

This is a replication of training and inference based on the provided details in the paper ( arxiv.org/abs/2412.06769 ) by using the same dataset for math as in the paper.
December 22, 2024 at 2:18 PM
Reposted
Honestly? It rocks that we just live in a world where a non-trivial number of people who vote are now just completely detached from reality.

Actually just living in fantasy worlds with alien drones and AI dogs and shit. Amazing.
December 16, 2024 at 12:40 PM
I stolen this from x @kylekulinskifromx.bsky.social
December 15, 2024 at 11:20 AM