[fs111]

> Maybe I'm being naive but how can a company that replaces a whole layer of management and overhead of traditional Taxi companies by an App not reach profitability?

They employ more than 2000 Engineers, many of them in the Bay Area. Those cost easily more than any manager in a Taxi corporation..

[Arizhel]

Yep. Amazing how you need 2 thousand engineers just to make an app (which is already made) and run some servers (which are probably hosted on AWS anyway).

[Balgair]

Well, the greyball program took some real coding and testing out to implement as it was reported. From the point of view of 'if you see one cockroach scurry on the floor, your walls infested with them' then we can make an educated assumption that there are more 'secret' projects they may be working on to further obstruct justice (highly dependent on location). The question is then: How many coders work on the core app, how many worked on the greyball section, how many worked on the self-driving cars (auto-automobile is maybe a better term for these vehicles, but not by much), how many does that leave left? Take that answer, assume that the greyball is average the number of coders on a 'secret' project (very very dubious to do this, but what other data do you have?), and then you have an estimate on the number of other 'secret' projects they have going on and what their abilities are as compared to your cadre of coders. Then try and steal them if they are doing better than your folks, now is a good time to invest in their coders before the crash and rush for them happens.

[sosborn]

> just to make an app (which is already made) and run some servers

I assume they want to do more than let the product stagnate, not to mention the resources they are putting into driverless vehicles.

[Arizhel]

You don't need thousands of engineers to keep a mobile app up-to-date, and the driverless vehicle thing is a boondoggle as there's already other companies doing that work, which are much better at it (Google, GM, etc.), and don't need tons of VC for such a risky endeavor.

[sosborn]

> there's already other companies doing that work

In your opinion, is work not worth doing if those mammoth companies are pursuing it?

Also, you say it is a risky endeavor. Would you argue that the driverless vehicle is not an inevitability? I would certainly argue that betting on Uber to come out on top is risky, but that doesn't mean there isn't money to be made.

[Arizhel]

>In your opinion, is work not worth doing if those mammoth companies are pursuing it?

Depends on what it is, and what kind of resources you have. In this case, you're talking about a massively difficult task (between the technical and regulatory aspects and the safety-critical nature), and at least two mammoth (i.e. well-funded internally) companies are already working on it. What exactly is a company with zero expertise in the field (a mobile app? Please), with absolute reliance on VC funding, likely to bring to the table? A lot of risk, that's what.

Yes, I think driverless vehicles are probably an inevitability, but I wouldn't look to Uber to come out on top here, any more than I expect Tinder to come out on top in asteroid mining.

[freehunter]

Saying that if they're not working on self-driving cars, then they must think self-driving cars aren't worth pursing is a non-sequitur. Every Uber I've been in has had music playing, why don't they buy some satellites and build a SiriusXM competitor?

Uber isn't even profitable with their core business of being a not-quite-taxi company. They quite obviously don't have the money to build their own self-driving cars, too.

[iamatworknow]

I wonder if they were just overzealous in hiring and don't actually need all those employees, but now can't let them go because of the impression that would give to the industry/equity holders. Large staff cuts usually do not instill confidence in investors, even if that staff literally isn't required.

[GhostVII]

Their backend consists of hundreds of microservices that are frequently breaking, so it's more complex than just an app and a basic backend on AWS. Also they have been doing a lot of massive refractors in the last couple years, so they need lots of people for that.

[passiveincomelg]

How much of that is self-imposed complexity from chasing the hype of the week?

Also:

1. Software is all about automation and increasing efficiency!

2. Let's hire thousands of programmers!!1!12!

Empire building much?

[GhostVII]

I would imagine most of the complexity is from handling a massive number of requests from around the world, and needing all of their data to be backed up and secure. They have their critical data backed up to like 10 different servers, when your dealing with scale like they are things get complex really quick.