News for hackers. Stuff that matters.
Re:Netflix never used its $1M algorithm (2012) by shermantanktop on Wednesday January 10, 2024 @04:27PM (#38943416)
Everything that Netflix says or does externally wrt to tech appears to be for this purpose, including open-sourcing some things. It is all catnip for the HN crowd to come work for them.
Re:Netflix never used its $1M algorithm (2012) by mrguyorama on Wednesday January 10, 2024 @05:00PM (#38943914)
Netflix pays in the same ballpark as other FAANG companies but the product is so much clearer (stream people movies and shows, maybe recommend them something to watch) and the complete lack of grey ethics like in AdTech or a Monopoly like Microsoft is the real catnip.
Re:Netflix never used its $1M algorithm (2012) by ketzo on Wednesday January 10, 2024 @06:33PM (#38945146)
I think it’s worth noting that actually, they pay the most of the FAANG companies, and they pay all cash, no RSUs. They also have some extremely generous benefits - one that comes to mind is 6mo fully-paid parental leave.

Netflix compensation is fucking bonkers.

Re:Netflix never used its $1M algorithm (2012) by ipaddr on Wednesday January 10, 2024 @06:46PM (#38945286)
Many things have come out describing it as toxic culture on hackernews. New employees last less than a year.
Re:Netflix never used its $1M algorithm (2012) by paulcole on Wednesday January 10, 2024 @07:18PM (#38945569)
> New employees last less than a year

Isn’t this ideal? Find out if it’s for you quickly? And if it’s not move on to something that you actually will like doing?

Re:Netflix never used its $1M algorithm (2012) by ipaddr on Wednesday January 10, 2024 @09:46PM (#38946823)
Not for new employees. If you knew you were going to be let go 9 months later I'd rather take a different offer.
Re:Netflix never used its $1M algorithm (2012) by paulcole on Thursday January 11, 2024 @08:23AM (#38951874)
But you don’t know that lol.

If you knew you were going to stay there 9 years and make a tractor trailer load of money and love the culture you’d take the offer.

If you’re not sure which is more likely and you really like the maximum upside then you should take the offer and get out in 9 months if it’s not for you.

Re:Netflix never used its $1M algorithm (2012) by ipaddr on Thursday January 11, 2024 @01:31PM (#38956508)
In this case we do know on average you are out in 9 months. Do you gamble you can somehow make an important friend or you will out political influence some of the 10 year vets; maybe you take the gamble. For the average person you would be better offer taking the other offer.
Re:Netflix never used its $1M algorithm (2012) by paulcole on Saturday January 13, 2024 @08:11AM (#38979615)
Well we know that in this case because you just made it up and added it to the story lol.

And yes, if I knew that the average new hire at Netflix lasted 9 months, I’d still probably take it. I think I’m much better than average and I’d want to test myself at a place that is challenging to work at. If I’m good enough to get a Netflix offer I can probably get another offer in the future.

The average person should do what the average person wants but the upside of a long career at Netflix seems pretty darn high and might make an average person an above average person.

Re:Netflix never used its $1M algorithm (2012) by diab0lic on Wednesday January 10, 2024 @09:17PM (#38946604)
The HN take is highly sensationalized. It’s a great place to work with fantastic work / life balance, great pay, and friendly coworkers. I was there for nearly a decade and would never describe it as toxic.
Re:Netflix never used its $1M algorithm (2012) by ipaddr on Wednesday January 10, 2024 @09:41PM (#38946783)
If you have been there for 10 years you might be part of the culture problem.
Re:Netflix never used its $1M algorithm (2012) by shermantanktop on Monday January 15, 2024 @04:33PM (#39006379)
Really? Do you always endorse the external perception/criticism of places you have worked for a while?

I know I don’t, having worked at places where I barely recognized the company from the ridiculous things people said about it on the internet.

Or maybe I’m the problem, wherever I go.

Re:Netflix never used its $1M algorithm (2012) by cybrox on Wednesday January 10, 2024 @05:11PM (#38944077)
They use a lot of the same concepts as AdTech but only for their own content without shilling for everyone who is willing to pay.
Re:Netflix never used its $1M algorithm (2012) by Cyphase on Wednesday January 10, 2024 @11:48PM (#38947750)
I think that context makes a reasonable difference for some people.

(The following is a tangent.)

To put it another way, using a slightly different kind of situation, I don't want BigCorp's AI watching everything I do to "serve me better", but I'm very interested in the idea of a local AI that watches everything I do and helps me be N times more efficient, while I maintain full control. I don't care if the local AI uses the same data-gathering, data analysis, etc. concepts as BigCorp's AI to accomplish the goals I want it to accomplish.

Caveats include:

- The local AI is Skinner-boxing you or similar. This isn't about the AI having conscious malicious intent; it could be a design defect. Also, a local, offline, open source video game can still be addictive.

- The risk of the system being compromised / manipulated, and all that data on you being exfiltrated.

Re:Netflix never used its $1M algorithm (2012) by lulznews on Wednesday January 10, 2024 @05:32PM (#38944373)
Which is odd given their app is highly trivial.
Re:Netflix never used its $1M algorithm (2012) by jamesrr39 on Wednesday January 10, 2024 @06:27PM (#38945071)
Hmm. I can't think of any time I've thought the app was really missing something, or didn't work. In fact, I always thought the app was pretty smooth and "just worked".

By comparison, several years ago, I was regularly streaming Netflix content without any trouble. One day I streamed a show on the HBO app on mobile data and it ate up almost all of my month's data usage. I guess it's behind the scenes work, and not so exciting, to e.g. stream a smaller resolution for a mobile phone, but in moments like that you realise there are a substantial amount of things to think about when making such an app.

Re:Netflix never used its $1M algorithm (2012) by amaranth on Wednesday January 10, 2024 @10:21PM (#38947103)
If you live in the US there is a good chance that was your phone company throttling Netflix and the app automatically switching to 480p to adjust to the reduced bandwidth. They probably do that more broadly now but when they first started it was only Netflix.
Re:Netflix never used its $1M algorithm (2012) by jamesrr39 on Saturday January 13, 2024 @04:10PM (#38984598)
I don't live in the US, but how would this work? Isn't the video sent over SSL? In that case how would the phone company know what to send back?

I just assumed that Netflix had different video formats on its' servers, ready to go, and when the app requested the video it also sent the screen size/resolution to get the smallest file size, that would also look good on my phone. If that was 480p then all well and good, it was fine to watch.

It was a while ago now, but IIRC the amount HBO used was pretty horrendous, like a GB for 30 mins or so of video, so I guess they just downloaded a big file and let the player on the phone handle that.

Like a say, a few years ago now so the situation may well be better with HBO now.

Re:Netflix never used its $1M algorithm (2012) by seanp2k2 on Wednesday January 10, 2024 @08:07PM (#38946046)
Building one of the largest CDNs on the planet is trivial? https://openconnect.netflix.com/Open-Connect-Overview.pdf

Changing the way Hollywood and content licensing works is trivial? https://www.hollywoodreporter.com/lifestyle/arts/how-netflix...

Yeah, it's not rocket science to build a streaming service using AWS or GCloud lego blocks in 2024, but remember that Netflix started streaming in 2007.