3 ms·

Could you elaborate a bit on where the bulk of the technical effort is being applied at Netflix these days, aside from recommendations and improved scaling? Fro

Could you elaborate a bit on where the bulk of the technical effort is being applied at Netflix these days, aside from recommendations and improved scaling? From the point of view of the user, the product has seemed fairly static for the past several years. But obviously there must be a lot going on behind the scenes. I assume the broad buckets of continuous effort are "get the right content to the right users", "add more content" (which may involve some engineering support via e.g. analytics), and "scale it up to effectively reach more users". It seems like Netflix has fully shifted from a chaotic "zero-to-one" innovation state to a more mature "continuously aggregate marginal gains on top of a solid foundation". Is that a fair characterization? The only conceivable "problem" with the above from the engineer's POV is that (1) individual contributions might feel less exciting and (2) it becomes progressively harder to find ways to move the needle. Though I would assume this is a problem at virtually any mature technology company. On the bright side, a marginal gain applied at a massive scale probably feels pretty rewarding.

city41 6y ago
I can definitely say a lot of focus and innovation is happening on the studio side of Netflix. This is where tools for casting, managing the logistics of filming, budgeting, etc are being built in order to scale up Netflix's ability to create content. I spent the last 2.5 years of my time on this side. https://jobs.netflix.com/teams/studio-technologies https://jobs.netflix.com/teams/studio-technologies My first 3 years were on the consumer side, working on the TV client. I'm a bit more out of touch here. But I'd still say interesting engineering is happening here. I agree sometimes it is "marginal gain at massive scale", but that can be really satisfying and challenging in its own right. There are many different teams focusing on a lot of different things here. The consumer side is pretty much entirely driven by A/B tests. So teams might take on a big idea, but it can be hard for those to beat out the established norm sometimes. That may be a factor in why you feel the product is "static". Also the wide range of devices and internet connections can segment engineering. Many efforts apply across the board, but other projects might only apply to say high end game consoles for example. Regarding "get the right content to the right users", I don't have much experience with that. That's more in the domain of backend teams. I do know they also do A/B tests to guide innovation, but that's about all I can confidently say there.
- rebelos 6y ago
  This was really informative. Reflecting on it, it makes perfect sense to look for software economies of scale in content production with internal applications and platforms, especially as that becomes an increasingly key differentiator in the streaming space. Thanks for taking the time to respond.

city41 6y ago

I can definitely say a lot of focus and innovation is happening on the studio side of Netflix. This is where tools for casting, managing the logistics of filming, budgeting, etc are being built in order to scale up Netflix's ability to create content. I spent the last 2.5 years of my time on this side. https://jobs.netflix.com/teams/studio-technologies https://jobs.netflix.com/teams/studio-technologies My first 3 years were on the consumer side, working on the TV client. I'm a bit more out of touch here. But I'd still say interesting engineering is happening here. I agree sometimes it is "marginal gain at massive scale", but that can be really satisfying and challenging in its own right. There are many different teams focusing on a lot of different things here. The consumer side is pretty much entirely driven by A/B tests. So teams might take on a big idea, but it can be hard for those to beat out the established norm sometimes. That may be a factor in why you feel the product is "static". Also the wide range of devices and internet connections can segment engineering. Many efforts apply across the board, but other projects might only apply to say high end game consoles for example. Regarding "get the right content to the right users", I don't have much experience with that. That's more in the domain of backend teams. I do know they also do A/B tests to guide innovation, but that's about all I can confidently say there.

rebelos 6y ago
This was really informative. Reflecting on it, it makes perfect sense to look for software economies of scale in content production with internal applications and platforms, especially as that becomes an increasingly key differentiator in the streaming space. Thanks for taking the time to respond.