Software Gurus...estimate size of Tesla software development team

[qdeathstar]

42.

 

[argon2018]

Outside my field of expertise....can someone with some insight estimate the team size Tesla likely has in place to develop their software? I'd like to EXCLUDE FSD and Energy products (ie Powerwall) from this exercise, if possible. Please consider the following:

• They are developing software for 3 vehicles (MS,MX, M3)
• They are performing bug fixes, functionality upgrades and regression testing (plus whatever else software engineers do)

I'll let the cat out of the bag on my intentions here. With the relatively slow progress on some bug fixes (black back up camera, or loading errors with media, etc) and expected enhancements (ie watch Sentry/Dashcam footage in-car), I suspect the team is relatively small compared to what it ought to be....but let's talk numbers before I pass too much judgement

It takes apple to release one year to release an OS that is buggy with barely anew features. They are 50X bigger than Tesla. It’s unbelievable the rate Tesla released new features and for free considering how much smaller they are. I think you should be admiring rather than pass judgement

 

[focher]

Some have added a lot more that I didn't consider. I agree it's in the hundreds.

For those not in the Bay Area or in the market for developers, Tesla is very well known not to use contractors overall (see previous Musk move to eliminate them - from the company, not existence) so I doubt they have anything more than a handful of development contractors.

 

[camalaio]

It takes apple to release one year to release an OS that is buggy with barely anew features. They are 50X bigger than Tesla. It’s unbelievable the rate Tesla released new features and for free considering how much smaller they are. I think you should be admiring rather than pass judgement

Bigger definitely does not mean more output. I've worked places with 200+ engineers and developers that could have accomplished much of their work with, say, 30 very well qualified people (some of which they already had, of course). In fact, I've worked in much smaller teams (<20) that had shockingly high output with great quality releases. The companies with more developers are in an awkward spot where much of their operation needs to be managed a lot more, which becomes a very significant portion of software development at that scale. At that scale you also need to take testing much more seriously as components develop at different paces from each other (similar to management in many ways, really) with increasingly complex interactions between them.

That was basically going to be my response to this thread. They could have very few (50 or so) or they could have much, much more. It's really hard to tell just looking from the outside.

Like others mentioned, there's also operational "internal products" that require development resources that we never directly see. Also very common of growing organisations versus smaller startups, and another source of needing more developers.

Source: Am developer, tester, & tech lead in various previous roles.

 

[stonelance]

I know of one person in my group\company that works there now, and I know they recruit pretty actively (I used to get pestered a bunch, but not as much any more) so that probably means it grew quite a bit and is large enough to have a pretty regular amount of turn around. I would guess 200-400.

I think the problem with slow updates is not a people-power issue, but a prioritization and scoping issue. I'm sure people on the team want to do everything, or at least all want to do different things, and often companies\managers\design are not good at focusing that energy. I would not be surprised if some of the initially "half-baked" features (in my mind) were the result of "hack time" and personal interest.