ZeroTalent

yup, i haven't had my coffee yet when I was writing it either :)

This. The same thing is happening with Tesla stock. Since it's such a bit part of the S&P500, so half of retail investors buy it in their basket of portfolios and 401ks at trades at 10x it's worth on paper by looking at fundamentals.

They just did this:

"As an added security measure, we’re requiring that all customers choose a new password unique to 23andMe. To proceed, please reset your password."

I did my test over 10 years ago and lost access to that email.

NICE.

Immigrants commit a bit less crimes than the locals in France.

10x is purely on payroll, which is not that great. there are a lot of other things you have to account for. There are companies that do 100x or more.

the other commenter said it better: https://news.ycombinator.com/item?id=43436688

Imagine buying something for $100-150M and having it disabled...

Many SOTA apps are, unfortunately, only for Apple M Macs.

Look into https://superwhisper.com and their local models. Pretty decent.

KYC is easy to beat on most platforms. look into stealer logs, as in the cybersecurity terms— you get the victim's cookies and passwords and often can bypass 2FA. Often, they have enough information to beat KYC.

I would say curation and editing are much more important than creating the art itself, but that might be a very unpopular opinion.

The people who bypass the paywall wouldn't pay anyway. It's just a marketing tactic. This + SEO.

UIs in React are very verbose. I'm not saying this is running 24/7.

Cursor is not SOTA. It's just popular. Look into alternatives. CLine, Augment Code, etc.

I'm no expert, but I have a feeling that the poverty line is not correct.

I know this is anecdotal, but I visited nine cities in India last year (including New Delhi), and the sidewalks were full of homeless people. At night, it was literally hard to walk through some areas because people were sleeping everywhere in the streets.

I've never seen anything like that in the US or Europe.

I might have a huge blindspot.

The thing is, when you're doing frontend, a human programmer can't write 4,000 lines of React code in 1 hour. A properly configured LLM system can.

This is why I wouldn't hire a person who doesn't know how to do this.

10x more effort once, 10x faster programming forever. Also once you got examples of the rules files, LLM can write most of them for next projects.

The tests are the program

I don't touch Brave because of the crypto thing and founder's affiliations.

It killed my battery. tried for a month.

"it was just a prank, bro!"

NVIDIA is one of the most undervalued companies in the SPX looking at fundamentals, even disregarding what they are planning next (subscription services for CUDA, and humanoids, etc).

Take a look at their last quarter's income statement graph: https://i.imgur.com/mQwZ5o4.png - Once in a few years I see a Sankey graph looking like that. And it's only growing over the last 10 years.

I've played around a bit with ibis for some internal analytics stuff, and honestly it's pretty nice to have one unified api for duckdb, postgres, etc. saves you from a ton of headaches switching context between different query languages and syntax quirks. but like you said, performance totally depends on the underlying backend, and sometimes that's a mixed bag—duckdb flies, but certain others can get sluggish with more complex joins and aggregations.

polars cloud might have an advantage here since they're optimizing directly around polars' own rust-based engine. i've done a fair bit of work lately using polars locally (huge fan of the lazy api), and if they can translate that speed and ergonomics smoothly into the cloud, it could be a real winner. the downside is obviously potential lock-in, but if it makes my day-to-day data wrangling faster, it might be worth the tradeoff.

curious to see benchmarks soon against dask, ray, and spark for some heavy analytics workloads.

Glad you mentioned Pi Unleashed. That tiny pitiny.c program is legendary in its own right. It really highlights the minimalism—squeezing out every bit of performance and precision.

CUDA is becoming more critical, not less, every day. Software developed around CUDA is vastly outpacing what other companies produce. And saving a few millions when creating new models doesn't matter; NVIDIA is pretty efficient at scale.

I don't know if you've heard, but NVIDIA is about to add a monthly payment for additional CUDA features and I'm almost certain that many big companies will be happy to pay for them.

> But most models will run or retrain fine now without CUDA.

This is correct for some small startups, not big companies.

No, because there is no CUDA. We have fast and cheap alternatives to NVIDIA, but they do not have CUDA. This is why NVIDIA has 90% margins on its hardware.

This is called lamphshading.

Lampshading is when someone acknowledges a flaw or controversial aspect upfront, often humorously or ironically, to deflect criticism or give the appearance of humility or openness.

Saying something like "A home for poorly researched ideas" while actually propagating extremist, controversial, or misleading content is a tactic intended to: disarm critics, avoid accountability, and create plausible deniability.

They will never let Poland go because it's a buffer for Germany.

Personally, I prefer that. I don't want to talk to people, usually. I want to go to the contact, see what communication forms we have in common, and send them a message.

I bet that they A/B tested this on users and this is not just a random change.

That's the problem for me too. Either too big or too small.

Need to buy 3rd party silicone inserts.

I'm gonna plug in my latest rant on why it's not worth incubating at Y Combinator: https://8figures.substack.com/p/exposing-the-accelerator-why...