4 ms·

agree with your comment Also, a genuine question Are Amazon, Twitter, Google, Facebook really doing any very technically amazing things? mostly they are buil

agree with your comment Also, a genuine question Are Amazon, Twitter, Google, Facebook really doing any very technically amazing things? mostly they are building platforms to collect more and more data and spy more and more that's hardly cutting edge stuff that needs the world's best engineers perhaps Facebook is beginning to realize this

Mehdi2277 6y ago
Google outpublishes most (maybe all) universities if we restrict to AI research. Not particularly surprising given they have several hundred AI researchers and likely over a thousand (deepmind alone is a couple hundred). They also do interesting distributed research. Multiple landmark papers come from (sequence to sequence learning, mapreduce, attention is all you need, etc). They have multiple major frameworks (tensorflow, angular). Facebook similarly has a ton of AI research and is one of the most prestigious labs you can work at. Talking to top AI researchers they see less and less of a motivation to choose academia over industry beyond teaching. Both facebook/google have a ton of freedom for their researchers. I'm focusing on ML just because that's the area I work in, but I know other research happens too, Microsoft has some programming language theory people and I think both microsoft/google have quantum computing research.
VRay 6y ago
My first real software job was at a 10 person firmware consulting shop. Some guy from Microsoft got burned out and joined us, and it was mind-blowing. The dude barely came into the office. Even when he was there, he was half-aslseep at his desk 75% of the time, and he still singlehandedly got more done in a week than everyone else in the company combined. Some of us had been theoretically doing the same work for 5-10 years, and our company was well-respected in its niche as an expert on what we did, and we still just couldn't compare at all. 10 years later, I was burned out from working at my crappy FAANG job building platforms to spy on people, so I went into consulting on my own. Sure enough, it was a really similar situation.. I don't want to break any NDAs or sound like a jerk, but I was rolling in and doing things in hours that teams of my clients' best engineers had been failing at for months at a time. You could grab just about any mediocre nobody from a FAANG company and drop them into any tech company somewhere else that normally pays sub-six figures, and they'd be the top performer there. I think it's something that can be taught, and maybe we'll have a tech renaissance now where everyone goes back to their comfortable hometowns, but there's a very real skill differential between the big tech hub cities and the rest of the planet. (And yes, I'm including other countries where even super hard-working geniuses make $35k a year or less.. because a lot of those geniuses eventually tend to either make their way to the USA, or start their own successful companies. If I found out I could eventually make $3.5 million/year by learning Chinese and working my way over there, you'd better believe I'd get it done)

Mehdi2277 6y ago

Google outpublishes most (maybe all) universities if we restrict to AI research. Not particularly surprising given they have several hundred AI researchers and likely over a thousand (deepmind alone is a couple hundred). They also do interesting distributed research. Multiple landmark papers come from (sequence to sequence learning, mapreduce, attention is all you need, etc). They have multiple major frameworks (tensorflow, angular). Facebook similarly has a ton of AI research and is one of the most prestigious labs you can work at. Talking to top AI researchers they see less and less of a motivation to choose academia over industry beyond teaching. Both facebook/google have a ton of freedom for their researchers. I'm focusing on ML just because that's the area I work in, but I know other research happens too, Microsoft has some programming language theory people and I think both microsoft/google have quantum computing research.

VRay 6y ago

My first real software job was at a 10 person firmware consulting shop. Some guy from Microsoft got burned out and joined us, and it was mind-blowing. The dude barely came into the office. Even when he was there, he was half-aslseep at his desk 75% of the time, and he still singlehandedly got more done in a week than everyone else in the company combined. Some of us had been theoretically doing the same work for 5-10 years, and our company was well-respected in its niche as an expert on what we did, and we still just couldn't compare at all. 10 years later, I was burned out from working at my crappy FAANG job building platforms to spy on people, so I went into consulting on my own. Sure enough, it was a really similar situation.. I don't want to break any NDAs or sound like a jerk, but I was rolling in and doing things in hours that teams of my clients' best engineers had been failing at for months at a time. You could grab just about any mediocre nobody from a FAANG company and drop them into any tech company somewhere else that normally pays sub-six figures, and they'd be the top performer there. I think it's something that can be taught, and maybe we'll have a tech renaissance now where everyone goes back to their comfortable hometowns, but there's a very real skill differential between the big tech hub cities and the rest of the planet. (And yes, I'm including other countries where even super hard-working geniuses make $35k a year or less.. because a lot of those geniuses eventually tend to either make their way to the USA, or start their own successful companies. If I found out I could eventually make $3.5 million/year by learning Chinese and working my way over there, you'd better believe I'd get it done)