The painful reality? This car can’t be sold in America. The US imposed a 100% tariff on Chinese EVs in May 2024. That means this $42K Xiaomi SU7 Max would cost $84K at the border before a single dealer markup. 660 horsepower. 0-60 in 2.8 seconds. 320 miles of range. Modular

Justin Bieber sounds just like a product manager 😂😂

The hackiest code you can write is obviously a website scraper where you actually depend on things not changing.

Life hack:

6 years of thinking I can make a better portfolio. 1 year of actually building it. My new portfolio is LIVE 🚀 Please, don't break anything! 👉 bruno-simon.com

There are competing views on whether RL can genuinely improve base model's performance (e.g., pass@128). The answer is both yes and no, largely depending on the interplay between pre-training, mid-training, and RL. We trained a few hundreds of GPT-2 scale LMs on synthetic

I still got his room running here: room.becreatif.ai A little bit custom, but he’s amazing.

I’m looking into the react2shell vulnerability and there’s a proof of concept to check so trying to understand what’s at stake here. There are some government websites that are very much vulnerable per the look of it.

Here it is running a shell command on a vulnerable version. Happy researching!

Not certain why people still build images and place secrets in it, what’s the fundamental benefit over supplying it when running the container. It’s a security risk, just stop.

Headers are capped to 8kb or so idk. But if you feel like extending this you can pipe results to robots.txt just to be funny.

And this is how I found out that my laptop was accessible over the internet. Interesting attempt.

Goal is to be able to pick up from any device and continue where you left off.

math bug is the worst bug ever! check your math before doing calibration bilinear interpolation collapses noise variance (1.0 -> ~0.25). Use NEAREST neighbor instead! 2nd pic is nearest 3rd is bilinear it's much more in distribution now even without calibration training