Agile development at Tesla | A conversation with Joe Justice

On March 12th, 2021 we had Joe Justice a second time as our guest to talk about his work at Tesla. Joe worked as a consultant for Tesla and shares his insights as well as the agile hardware development and working habits at Tesla.

Transcript: Agile development at Tesla

Sohrab:
We're very excited to have Joe Justice back with us. He was one of our speakers at the very first agile100. And now, for our ninth edition, he's back. In the meantime, he's been at Tesla, a pretty interesting company, and we're going to spend most of today's session talking to Joe about what he learned there, right? So, Joe, welcome back. Zuzi and I are going to do the interview with you together. We're going to grade you. We've given you a warning. Now, tell us a little bit about yourself. What's happened since we last saw you here, which was May 2020?

Joe:
When the pandemic started, I was traveling around Europe doing Scrum Master training and Product-Owner training, all in person. All of them were in different European countries. And then the lockdown came. I was in Switzerland teaching a course. I took off right when the flights were restricted and had 100% cancellations. All in-person in-house training was cancelled. My entire 2019 calendar year was cancelled. And I thought, "What am I going to do when we find out what this pandemic means?" And I said, "Tesla is changing the world, and I'm convinced that they're the future of work and they're the next generation or the next, next, next generation of what agile should be. Is that right?"

Well, I have time. All my classes are canceled. I'm going to apply. And I've applied to Tesla, SpaceX, The Boring Company, OpenAI, and Neuralink, in all kinds of positions: Project Management, Leadership, Welding, Shop Floor Management, Supply Chain Work, Composites, Programming. I had an interesting background. I was able to apply for most of the positions. Before that, I was a developer at Microsoft. I previously worked with Bill Gates in evangelism at the Bill and Melinda Gates Foundation. I also worked at a high level in project management and was an executive and business consultant for many Japanese companies. So I was able to apply for most of the jobs and thought, "This would be my pandemic job. I'm going to learn as much as I can."

Now, at Tesla, there are 12-hour shifts. Every day it's 12 hours. So there's overlap with the next shift, and operations run around the clock. And when we agilists talk about shared ownership, there's no way around it if it's a continuous 24-7-365 operation. So I would call it radical shared ownership. It's like XP, it's all there. So there are these 12-hour shifts, but I would come in an hour early because I'm an agilist who wants to figure out the process that really makes the world more awesome. So I would come in an hour early and take notes on the previous shifts, new product launch, new product development, sourcing, supply chain management, manufacturing, assembly, supplier management, logistics, and fill out a diary. And then I stayed another two hours and went to the department where I had worked that day.

That's not actually a department, that's the area that the stuff came from that I worked on. So I followed the product and ended up going through all of Tesla headquarters in Fremont, California, and most of SpaceX headquarters in Boca Chica, where Starship is made, and SpaceX headquarters in Hawthorne, California, and took notes for two hours after the shift and one hour before the shift. Then Tesla specifically allowed me to teach Scrum or Agile on the two days off. I often had two days off, the way the structure worked, and I taught.

However, I got exhausted pretty quickly, and I'm really fit. I'm a very fit person. But that only lasted so long. It was pretty exhausting, but I learned a lot. And I'll try to share everything I can without violating my non-disclosure agreement.

What was your first impression of Tesla?

Zuzi:
So let's start at the very beginning. What was your first impression when you started at Tesla? When you walked into the office, what was your feeling?

Joe:
Well, I can't remember exactly. Let me see if I can gather my thoughts. There's a movie from a long time ago, I don't know, 15, 20 years ago, called Starship Troopers. There's a scene where the protagonist and his friends join the Space Marines or the Space Army, and they get all this awesome gear and everything is totally futuristic. And yet it's really lean, maybe that's the right word. It feels fast-paced and yet it's super futuristic and exciting. That's exactly what it's like. I felt like I just signed up for Starship Troopers. I mean, it started as soon as I walked in.

The first thing that happened, and luckily it's already leaked out so I can talk about it. There's a lot of things I want to say but can't. This one I can. The first thing we all got was the employee handbook. It's called the anti-handbook handbook and it's four pages long. And that was it. And somebody else passed it on. I recommend everyone read it. It's the new agile manifesto. It's the agile manifesto if you want to make billions. So it's the manifesto that works. And it says, "You have full responsibility. If you see an opportunity, it's your responsibility to fix it or take advantage of it. You can do anything you want. You can talk to anybody you want. And if anyone tries to stop you, they're fired immediately." They didn't write the "fired on the spot" thing. Elon said it. So I get to say that too, Elon said it. He says, "And if anybody tries to stop you," Elon says, "you can talk to me. And if anybody tries to stop you, they're fired."

And really, from the moment I got in there, I could buy a robot just like anybody else. I could buy land. I could build a factory. I could try to sue another company. You have access to billions of dollars, no matter who you are, as soon as you walk into the company. And you can go wherever you want, and nothing is hidden from you. I then navigated through the hierarchy tree, because they use some.... they're all custom apps. As soon as you sign in, your phone is loaded with apps, and you do everything through apps. Everything is instant, chat. You don't need a laptop for almost everything. You just need your phone. And they don't give you a phone. It's just, "You have a phone, right?" And poof, you've got everything. And it's password protected and multi-authenticated. The security is pretty important. But it's whatever your phone is.

And you can navigate... Some of the software they bought, some of it they developed themselves. Some of the software is older, traditional software, almost like Microsoft Project. Of course, it's not used as much, but it's there. It's included in the package. So you can "navigate" through the hierarchy tree and you can see how the software tries to make it a hierarchy because there is no hierarchy. It's just flat. There's a video of Elon Musk saying, "This is where I sleep," and that's a sleeping bag in the paint shop, right next to where I worked most of the time. And yes, it's "go up the hierarchy tree," there's no "up." It's like, this is where they work. Okay, who's their "boss?" Oh, they're here. I just worked with them 20 minutes ago. Oh, who's your boss? Oh, Elon. Elon is sleeping over there and he was just working with me on this robot, or you know, in this area, in this area. Actually, most of the time I was there, Elon was at SpaceX, but he was sleeping right there.

And just to be upfront, I think that's why German manufacturing companies don't stand a chance. I've worked with the board of Bosch and the executives at BMW and met some of the executives at Porsche and Volkswagen, and none of them are going to sleep in a factory, and none of them are going to program a robot, and none of them are going to put glue on internal assemblies on an assembly line because, one, that's necessary, and two, that's how you learn. None of them will do that. They are "business people." They are "business people" and they will never be able to compete with Tesla. And they're going to die. And your stock tip for the day is that Tesla is worth about 100 to 1,000 times less than its projected total value. So I don't know if you're an investor. And investing is a gamble. We can all lose. But my advice for stocks is, yes, Tesla will go up 100 to 1,000 times in the next 10 years. So if you have some money that you can afford to wait, that would be a smart place to invest it.

How do 12-hour shifts work at Tesla?

Sohrab:
There's already a question from the audience, Joe, can you share with us how the 12-hour shift works? Does it apply to everyone, to every department? How do people adjust to it? And during the break, before people came in here, we've already started to go into some of these things. I think this is a good time to talk about it now, because you mentioned yourself that you didn't last long, in part because you didn't take the two days off. You were teaching on those two days. But in general, what you're saying sounds very exhaustive, doesn't it? Can you explain a little bit about that? In Agile, we talk about sustainable pace. How do you bring sustainability to this extreme way of working?

Joe:
I think that's super important for anyone interested in business performance and agile. There's a part of agile that says, "We want happy people writing software. That's part of agile. That's one of the undertones in the agile manifesto, and that's important too. Another part of agile is business performance, which is the part of working with the customer that contributes to business performance. And they coexist very well. So there are multiple perspectives.

I think I've learned that if you're doing a lot of work, it's got to be super fun. When Musk introduces Giga Berlin, he begins by describing the Rave Cave, the massive nightclub that is a key part of the factory. In fact, the Rave Cave is one of the most important features and structures of this huge factory. And this is not meant trivially. When you're working so intensely, it has to be what you would do for fun anyway. And if that's the case, that's one aspect of sustainable pace. And I think a lot of agilists get that. They made their scrum room, their team room, or their mob space really fun. They brought in sand, kiddie pools, fairy lights, and music. They try to design the space the way they would use it for fun anyway. And Musk understands that. So there's a lot of fun involved.

At SpaceX Boca Chica, where the world's largest rocket, the Starship, is being built, there's a bar, a tiki bar, a Polynesian-style bar atop one of the launch towers, and a restaurant made out of exploded rocket parts. And that's so cool. A lot of the people who work at SpaceX, Tesla and The Boring Company would go there anyway. If they could go anywhere in the world to have fun, they would go there anyway. So that means you don't go away from work to have fun. During the time you want to have fun, it makes sense to still be at work and talk to people about your productivity. So that's one method: make it absurdly fun or absurdly fashionable or absurdly luxurious, but make your work your ideal destination. So there's some power in that. And the second part...

Sohrab:
That's kind of like what Google has been doing all these years, right? Making the campus such a great place that people don't even want to leave. There can be downsides to that, especially when people leave to spend time with their families, right? But I'm sure you'll get into that, I just wanted to make the similarity to Google's approach.

Joe:
It's a very common idea in Silicon Valley. I'm not sure how many here have had the luxury of working in Silicon Valley. It's normal in Silicon Valley to have baristos and baristas who are mixing any coffee drink and trying to make it world-class, and really striving to be world-class 24/7. That's normal in a Silicon Valley company. Tesla does that too, and the baristas and baristos are very good. They make coffee the Australian way. They make coffee Parisian style. They make American café-style coffee, Italian, depending on what you like. They're trained by real elites and they're really trying to be world class, and they're good. They also try to serve Michelin three-star quality, and sometimes they succeed: free food every 3 to 4 hours, around the clock, hot. In that way, it's the best or almost the best food you can get anywhere in the world, and it's there. So you might as well stay there, as Sohrab says. And that's very normal in Silicon Valley.

Google's sushi restaurants were legendary. I'm not sure they still are, but they were legendary. They were fantastic world-class sushi restaurants on the Google campus, and there's a climbing wall and a post office and a dry cleaner and a hairdresser. So you can get those things done. We've been to the Microsoft campus, which isn't quite in Silicon Valley but has a similar culture, or Google or Amazon, etc. Starbucks is just like that. And so is Tesla. So that helps. That helps. I think I missed part of the commentary that you led us to, Sohrab.

Sohrab:
Yeah. How do you make it sustainable?

Joe:
Okay. Well, there's a second part and then there's the opportunity to get better. And if any of us have ever played a professional sport - I've never been a professional - or known someone who's played at the professional level, they know that it affects their diet and their sleep patterns. They say things like, 'Oh, I have practice tomorrow, so I'm going to bed early.' And they mean it, because they're professionals, right? That's a lifestyle choice, and not everybody wants to make that choice. Not everybody is going to be a professional. If you treat it like a professional sport and say, "I'm going to bed now because I have to work tomorrow," then it's game day for you at a company like Tesla or SpaceX. It's a game day. Every workday is either training, like in professional sports, or game day. So it's, "Okay, I'm going to go to bed on time, and I'm going to eat better. I'm not going to eat that because I know I won't sleep as well." And so you have to set your body. You have to treat it like a professional athlete.

And that also means that your support system has to be set up well. If you know a pro athlete or you were a pro athlete, your spouse and friends either have to help you and check the mail or mow the lawn because it's game day, or pick up the kids because it's game day, or you have to simplify your life so you don't have those things. You live in an apartment where you don't have to mow the lawn, and they scan your mail for you and email it to you so you don't have to pick up your mail in person. And your kids live with your grandparents because you're a professional athlete, right? And that's a lifestyle choice.

If you simplify your life or have a support system that allows you to be a professional athlete, any intense task, like professional sports, becomes more attainable or possible, but that's a choice that not everyone wants to make. If you do it and you enjoy it, yes, it's sustainable. That being said, I think there are some things that we agilists can bring to this new wave that will make it accessible to even more people, and I think that's really important.

I don't think I'm the only one who loved Agile Testing Days and Agile Testing Days Europe as a conference and ALE, Agile Lean Europe. That was probably the most fun I've ever had at agile conferences, and maybe the most I've ever learned, and that's an interesting and useful overlap. And I think Elon is trying to get the Musk companies to be more like ALE, Agile Lean Europe, and more like Agile Testing Days at work, but I don't think the Musk companies are there yet. I think they still have a lot to learn from the organizers of ALE, how to be super productive and learn a lot. I mean, ALE is an enormously... for those of you who were there. I think, enormously effective learning experience. You can teach Musk companies a lot more that Musk companies haven't learned yet. So I think we agilists have a lot to give back as well.

What else can Tesla learn from agilists?

Sohrab:
Yes. What would those things be specifically? Give us a couple of examples.

Joe:
Our Definition of Done and Definition of Ready, so you can engage yourself, is absolutely present in Tesla, but one of the things I need to do is bring a lot more DoR and DoD and Open Space, right? Open Space works well because everybody understands that there's a law of two feet and there's no shame in somebody leaving your session, or there's no shame in you leaving somebody's session. It helps to post these very simple, elegant, wiki-style object-oriented rules, and I did that a lot when I was running Agile@Tesla at Tesla, which reinforced the culture. So that's part of making the simple rules more explicit.

And honestly, I think Scrum works very well because it has very few rules. There are three roles, five events, and one activity. It gets more complicated, not less. It probably should be three events, but anyway, it's five plus one activity and then three deliverables. That's a pretty short list. I think we agilists have carved out such well-thought-out, research-backed, simple, easy-to-remember rules that many really high-performing companies don't have, which leads to burnout. Tesla has DoR and DoD, the definition of "ready" and "done" for each position, which makes it easy to get started. People can step into a whole new area of the company at any time, and we did.

The visual management could be better, but it's really good and in some ways much better than what most of us agilists have. But there are some Kanban flow WIP features that some innovative software teams have developed that I haven't seen in Tesla that would make visual management of work easier. Estimation or Planning Poker?. I didn't do dot voting. When I was at Tesla, I didn't see dot voting. It's a fantastic, simple agile practice. And it's not even necessarily agile, but agilists love it. If you do it well. There are a couple of agile coaching practices that would really help these companies.

Zuzi:
What was the hardest thing for you when you first started? You have your ideas, right? But the reality might be different. What was the hardest thing for you to digest?

Joe:
Well, there's the fun answer that you want, on the one hand, and the actual answer, on the other. And the actual... Joe: yeah, I'll give it a shot, yeah. The actual, concrete answer is... oh, man. That's it, isn't it? That's not the romantic answer. The hardest part was that the restroom was a seven-minute walk away and three stories up, where I was most of the time, and it was really hot, and you're wearing face shields, goggles, two layers of masks, and nitrile or plastic gloves, because it's COVID time, right? So you're kind of in a little space suit. And you're bouncing. So you've been working hard, right? I've been building thousands of parts for the Tesla Model Y, Model 3, S and X, developing new products, integrating new products, prototyping. So you're thinking harder than you ever have before, and you're also working physically.

In these companies, you're working full time. There are no desk jobs. There's no job where you're not constantly lifting 20 kilograms no matter what, right? So there's no job where you sit down. There isn't. But there's a lot of mental work. You just do it while you're physically working. So you've worked hard, and then you have to walk fast for seven and a half minutes, or even walk slow, and go up three flights of stairs to pee. And then you go pee because it's your only break, and you're like, "I'd really like a coffee. I don't have time." Because it's COVID time, I'm going to wash my hands for 60 seconds. So you wash your hands for 60 seconds and then you put your gloves on, the next layer, the gloves. It's like you're going back into the OR, right? You check your protective gear and then you rush back seven and a half minutes so you're there in time for the next group of robots. That's actually the hardest part. And here's the shortcut that's wrong and causes some of the problems. I don't think I'm allowed to say the next part because it's not published, but at least I'm allowed to say the first part. Maybe you can guess what I would have said.

The hardest part is that you have full decision-making authority and it's really tempting to say, "I just solved a problem that would have made me vice president at Toyota," right? It was such a big deal. If I was at Ford, I would have just won an award, and the CEO would have presented it to my group and my team, with our team name engraved in crystal, and we would have all gone out and had champagne, right? There would be these dramatic recognitions of achievement. Well, in a company like Musk Companies, that happens multiple times a day on almost every team because it really is that much more awesome. I'm not even kidding. How do you think the safest SUV in the world by far was built in less time than it takes BMW to change the shape of an air intake, and they still haven't fixed their navigation system? So it really is that huge. By the way, I love BMW. But there's not that recognition. There's no champagne party. There's no time. I mean, the next batch of robots and cars is here.

The hardest thing is the toilet was seven and a half minutes at fast walking, but the hardest thing is you expect, okay, we did it, but really, that phrase, the reward for hard work is more hard work, yeah, you have to pat yourself on the back because.... I mean, everybody's going to say, "Good job," but that's it. You're not going to slow down. There's not going to be a party where they shut down production because you were so great. That's what's expected. It's expected that you just did the best job in the world. So you have to congratulate yourself.

What does empowerment mean in a company like that?

Sohrab:
Joe, I have a question here. You mentioned empowerment, and I'll just ask that question briefly and then ask a follow-up question myself. Would you empower yourself to order new restrooms closer to the line?

Joe:
Yeah. That actually would have been the solution, and that's what I was working on when I finally stopped. So, yeah. That was also because Tesla's headquarters at that time would probably be in Giga Texas or Giga Berlin, but at that time it was Fremont, California, and that was a former Toyota and GM plant. So, yes, you can put stuff in, but there's a lot of thick concrete in places that you don't necessarily want. Technical debt, right? There was technical debt from a whole different waterfall method when Toyota and GM built the plant. And they just put the toilets in very inconvenient places. But we could have built portable toilets that were pumped or maintained by robots, and that's what I was working on, because that's super important.

Sohrab:
My follow-up question has nothing to do with the toilets, it has to do with empowerment. You mentioned that there is such a thing as extreme empowerment. In most organizations that I work in, and maybe it's true for you, Zuzi, and many others here in the audience, it's usually not enough to empower employees, especially if they come from an old culture, old structures, processes, and metrics. It takes a lot of coaching to get people to be more self-organized and self-managed. And even when they've done that, it still takes a lot of coaching from leaders to continually develop people, right? And how is that handled at Tesla? And maybe my assumptions that I just shared are completely wrong, looking at your experience at Tesla and probably other Musk companies, as you call them.

Joe:
Well, among agile trainers and coaches, consultants or professionals, I'm very unusual in that I've always taken the view that an agile transformation can take four hours or less. That's a very unconventional thought or point of view. Most agile consultants, trainers, coaches, and consulting firms talk about an agile transformation taking years, and that's in line with their experience. I don't think they're wrong, and I think for the methods that work well for them, that's right. I take a very unconventional stance, and my work history supports my unconventional stance. And maybe that gives me a unique perspective on it. At Tesla, it's been emphasized so often that you're 100% self-reliant that there's a certain inertia, right? There's a cultural inertia that it would be stupid if you weren't, that's what it feels like, and there's nothing that points in any other direction. There's nothing about annual reviews or annual goals or anything that points in a different direction, and I don't think coaching would help with that. I don't think it's necessary in a company like this.

In other corporate cultures where there's cultural inertia, it's a massive mindset shift because people's habits and behaviors are going to reinforce a less agile culture, and that means that everybody has to change their habits. And that will probably take a whole army of coaches leading by example. But if you already have an agile culture and you bring in some people with a waterfall, unempowered, or handoff mentality, it's going to blow out of them in minutes because there's nothing compatible with it. One of the reinforcements is that there's almost no place to sit down, I mean, you're doing, and there's clearly no handoff because you're going from the definition "done" to the definition "done" with the product. So the work enforces that it's unphased, and the work enforces that there are no handoffs. And you... everyone tests their own work, period, so there's no concept of handoff to test. And the visual flow indicators... how do I put it? There's no room for adulation.

What is the influence of work style and culture at Tesla?

Zuzi:
They have a different work style and culture, right? So do they have a different interview style as well? Do they prepare you or test you to see if you're a good fit, or what's the interview process like?

Joe:
So, the teams that you're going to work with first, and there's a radical movement, I mean, you go all over these companies, but the teams that you're going to work with, they interview you directly first, you know. So they'll interview you, which is a very agile practice anyway. They'll interview you and then they'll decide if they want you on the team. So that means the performance and quality of each team and even each individual, I mean, you guys work like a mob. It's very much mob work. It's similar to what Woody Zuill has been saying for a long time. It's very mob-like.

But it's still pretty clear what each individual can do, even though you're in that mob, because you're working so closely together. Because the team can ask you at any time not to come back, and that sounds very unhealthy, there's a lot of respect for people, a lot. There's a great deal of, "We've got each other's backs. We're in this together," shared responsibility. I would say you would have to do something very stupid to get fired right away, or something very dangerous or very disrespectful to get fired. But since you can be kicked out at any time by your teammates, there's a lot of security. I don't mean that to sound like there isn't. But it's not like you're waiting for a review and you can appeal or something. It's instantaneous. Everything is instantaneous. You're really doing your best to do your best every minute.

So I was under the impression that Tesla hires everybody, everybody, that's my impression. Yes, the team you're going to work with introduces you, but you're hired, that's my understanding. And if it doesn't work out in the first few minutes, they will ask you not to come back, but the extension of your interview is the real work. And so it happened that somebody that I worked with a lot, and we move around a lot, and our swarms and mobs move around a lot, but somebody that I worked with a lot was an MBA from MIT Sloan Business School, which is one of the more well-known business schools, and he was part of U.S. President Barack Obama's re-election campaign, as far as I know, in a senior position. So she is an extremely rare, possibly very talented, at least very educated person. Another person on our team, I think, was an ex-convict who used to be in prison and I don't think even graduated from high school. We all performed at an extremely high level as a team, and sometimes the convict was the better player.

So my impression of the interview was, "Okay, we want to get to know you so we can get started right away, not to check you out. We'll check each other out if we do the right job. So you're hired. I'm glad we got to meet you. Now let's see if we can really deliver."

Zuzi:
That's nice. I wish all organizations were like that, because then you'd have a chance, right?

Joe:
You'll get a chance, yeah. Everybody can, and I think everybody can get a second, third or fourth chance. So you can, I think, go in and out or go to Tesla, SpaceX, the Boring Company or some other company and come back. There's a culture of forgiveness that's very strong. If you're at work, they just want you to do a good job, and if you don't, they'll ask you not to work. And I think that's why it's so lean and why there are no desk jobs, because if you? have the power, you can build a desk and sit down and create five-year waterfall project plans. You can. And everybody around you would say, "You're not contributing to the mission right now," and you'd be asked not to come back.

Zuzi:
What is the percentage of women who work at Tesla? That's one of the questions from the audience. But I was also thinking about, "Is this for me or not?" So what is the percentage of women?

Joe:
In my experience, it's a little over 50%. I haven't seen an overall number, but my impression is that a little over half are women. And Tesla, specifically the Musk companies, was just voted the best place to work in the world for lesbian, bi, gay, transgender, LBGTQ, that group, for the fifth year in a row, maybe longer. They say there's no better place in the world to work. And I think that's because people really don't care if you've been a convict or you went to MIT Sloan or you're a man or a woman. It's all about, "Are you contributing to the mission?" The people who want to contribute to the mission can do that, and there's very little attention paid to whether or not you're good-looking or anything.

Sohrab:
So you're saying it's a real meritocracy, right?

Joe:
On a good day, yeah. I mean, no company is perfect, but I think it's closer than I've seen it before.

The Gigafactory in Berlin

Sohrab:
Christoph from the audience asks, Tesla is building a factory in Germany, the Giga Berlin, right?

Joe:
Yes.

Sohrab:
And what's your assumption? Are they going to be able to establish the same culture that you experienced? I mean, I read a report that they're hiring a lot of engineers and managers from Daimler and Volkswagen who have a very different background, and so I asked the question about coaching. What would you guess based on your own experience there?

Joe:
I'd like to ask some of the European agilists who like to work in the CET time zone, or are even local, to develop programs to prepare people from traditional top German automotive and manufacturing companies to have that mentality. I think Musk will solve the problem. I think Musk will solve it, and I think those people will get their act together. I think they'll have an easier time if they have a common vocabulary, if agilists who like to work on advanced training and people who like to.... Sohrab, your Scrum Academy would be a perfect example of that.... So you're thinking about joining Tesla because they're going to be hiring. They are already hiring tens of thousands of people. Here's a recommended induction program for you, and I think that's going to help a lot of people and make the onboarding process less traumatic.

That said, I don't think it won't work. I think the culture will remain, but I think people will be more comfortable coming in with some of the vocabulary and some simulations. I mean, many of us in this agile100 have seen the airplane game or the context switch loss game or the coin game or these agile games, and you've seen, when it's happened to you and you've maybe taught it, the gleam in your participants' eyes when they see how a piece of flow, when managed as a group, is dramatically faster or how production as a mob is faster than individual steps, for example, the airplane game, where the quality is immediately visible to everybody, test in the loop. A lot of people don't know that yet, and they'll find out quickly in Giga Berlin. But if they've done it before, they'll be less shocked. And I think that will help them.

Sohrab:
I mean, one good thing is that you guys are offering courses in the CET time zone, if I'm informed correctly. So everybody wants to learn about agile, especially in the world of hardware and terminology and examples from Tesla. I mean, you're welcome to attend one of Joe's trainings. I attended one of your hardware trainings a couple of years ago. We did that together at Bosch, and it was really very good. It opened my eyes to the extent to which you can apply the same techniques that we apply in software and services development to hardware, where in many cases - we've already had a session today - there's an assumption or a myth that it's not as maintainable when you apply agile practices.

What is the importance of the environment at Tesla?

So we've got 10 minutes left, Joe, and I wanted to go back to the issue of environment, because one of the points you made was that if you put people in an environment where that empowerment is constantly reinforced, that basically automatically leads to behavior change, right? I've always talked about organizational change being cultural change, and I've always said that culture is the collective mindset, and mindset is the habits that we exhibit. We can never get into people's brains, and this is where my son raises his hand. Okay, you get to see people once, but now go play. So that only happens here. Okay, one more. And then you go play too.

Joe:
Please let him play. We're agile101 now. Please let him play along.

Sohrab:
Yeah, or two. I have another one. But let's get back to the topic. I always say that you can never get into people's heads and change their neurons and habits, but you can always change the environment. And what I'm reading from what you're saying is that the environment is so dramatically different, with all its advantages and disadvantages, right? You're not saying that everything is perfect there, but that immediately causes people to behave differently. Am I understanding that correctly? I see you're nodding, but maybe you can elaborate a little bit more on that.

Joe:
The reason I think that an agile transition doesn't need to take more than four hours, and I understand that that's very controversial and contrary to the experience of many others, is what I just heard from Sohrab. And, Sohrab, please, play with your kids, too. If the work we do doesn't include family and pets, we're doing it wrong.

Sohrab:
You want me to pick up a package that was delivered to the neighbor. I can do that in 10 minutes, but not now.

Joe:
Yeah, okay. Sohrab: Okay. Other than that, the session is being recorded, so you can watch everything I say again later if you want. I mean, it's all good. I mean, it's all good. Yes. So, I started with my own business. A lot of agilists started by reading a book and taking a course, which works. It works. I took a much harder path. I started a company in an incredibly competitive industry. I started an automotive company that designs, manufactures, and tests its own cars, and that's really hard. I rediscovered a lot of agile practices out of necessity, and then later I learned some agile practices more formally to be more specific.

But what I noticed is that we did what we called a "build party" every Thursday, initially one car a week, then later several times a day, with new design. That was very agile, wasn't it? So agile means you design, analyze, build, test and deploy, right? If you do that once a week, a new car design, and then even faster, we become famous. We set four world records, and I'm really proud of that, and that experience changed my life. I noticed that the people who came to the Build Party were immediately productive and engaged, no matter who they were or what waterfall culture they came from, for example, a lot of executives came from Boeing, which usually has a waterfall culture.

The environment created behavior that made sense. And it was only safe because we had all these automated security tests. We basically had DevOps in hardware, with a huge regression and security suite, and we had something that still isn't as common as it should be. Those of you who wrote software before IDEs were really good, like command line development, which some people still do, didn't have color coding and autocomplete. It's still mostly like that with hardware. But it doesn't have to be that way.

The really fast companies, and my company, started introducing autocomplete and color coding when drawing CAD data to show that the feedback at design time is probably not strong enough, like software developers do in a mob, right? And in a mob where you have the driver, the navigator, and the other roles, you get that feedback from humans as well, but some of that feedback can be automated in context and shown to you at design time. That's basically DevOps at design time. That kind of support made it possible for people to be super agile and super nimble even with a very robust and practiced waterfall mentality and culture in that environment. And some of the longest-standing legacy mindsets that I've ever worked with were at least for a while in my car company in a build party, working with XP at an extremely high level, both hardware and software.

So, Sohrab, if I understood you correctly, I agree 100% with what you just said, that the environment can only allow part of the culture to make sense. That doesn't even mean that it changed their culture. It just means that only part of their habits apply, and those are the ones that are used, and the result is hyperagile.

Sohrab:
Yes, no, thank you. I also thank you for... I mean, the example you just gave, for those who don't know: It was WIKISPEED, right, Joe?

Joe:
Yeah, yeah. WIKISPEED is making houses now.

Sohrab:
And there's lots of footage, including videos on YouTube where you can see Joe doing some of these building parties, and I think you did some of them at Agile Testing Days as well, if I remember correctly. Yeah, yeah, cool. Let me take a look at that. Are there any other questions that I haven't seen? Maybe you have one, Zuzi. Ah, there's one. How is success measured per team?

Joe:
I think that's the problem. It's just expected to be best in class, constantly improving, and so, as I said, recognition is difficult or nonexistent, as Zuzi put it. So the benchmarks are all benchmarks for the product, and that's almost completely independent of you. So is Model Y lighter and safer? Is Model X faster and more beautiful? And beautiful is subjective, but they do their best to measure that continuously. And that's what they're measuring, and they're measuring it on huge monitors all over the place and with trends. So you can see, "Are we trending better? Are we not trending better?" And so you and your team can try to influence those things.

By the way, that's what's confusing investors and why Tesla stock keeps going down, it's also going up, but it also keeps going down because investors don't understand it. They're asking, "What's your profit model? What's your cost target per car and your sales target?" And the Tesla management team, Musk, Winterkorn, all of them, say, "We don't have targets." And they don't. All they have are metrics. That sounds wrong. All they have are trends. For example: How beautiful is the Model X? There's no target. It's just, "What are our combined measurements to judge how beautiful it is? Is it getting prettier?" And there's no target. "By the end of 2022, it will be 20% nicer." There's none of that because there's basically no management. It's just, "This is important. What can you do to make it more important?" And teams are measured directly on their contribution to it, such as efficiency and capital expenditures, but there are no goals. As a result, there's no individual measurement or management or even team measurement or management. It's only about the product. It's only about measuring the output.

Zuzi:
And I think that's really great. That's what we teach in all of our agile courses, right? That's so cool. I think we have time for one of the last topics, which is how long you stayed at Tesla and what makes you leave.

Joe:
When my son turned four, I was really tired because I had to work so many shifts. And I said, "Can I take time off for..." and anybody can take time off, but you want your team to know that you're not going to be there, "Can I not be there for my son's fourth birthday?" And it was when the heat pump was introduced from the Model Y into the Model 3, that was.... I mean, there's always something critical, but that was the critical thing at that time. The rear of the Model Y was cast in one piece at that time. And they said, "We're very busy. We'd really like to have you here, Joe." And I say, "He's going to be four years old. He's only going to do that once." And they said, "We'd really like to have you here." And I said, "Well, I guess I'll have to quit." And the management team... There's not really a management team. It's hard to say. I'll call it a management team so we can understand. They said, "Please don't leave." And I said, "I guess I have to resign. If it makes sense, I'll come back." And they said, "Please don't go."

So I resigned and went to my son's fourth birthday. And then, honestly, I slept for a week. I was super tired, too. It was also a decision of exhaustion. So that's how that happened. That was the summer of 2020, I was happy to be there when Tesla stock went up five times, and it's not like I was responsible for the stock going up five times, but I knew what the company looked like from the inside when that happened, and so I'll know my whole career what it's like to be in a company that has such dramatic, publicly recognized success. So I know. I was there leading the agile effort when it happened. Yep, but that was the reason I left.

What are the lessons learned from your time at Tesla?

Zuzi:
What's the one lesson you want to share with us?

Joe:
Man, can somebody please tweet me how to say this in a friendly way, because I'd really like to. But the thing that I've definitely learned is how to say it kindly and lovingly, because that's the goal. But the real learning, the only way I can say it now, is that there's no excuse, it's possible. Do you want to change the world or not? And you can. It is fully possible. It is fully possible. We already know how to do it. That's what the agilists have been saying for a while. They're right. Do you really want to do it? There's no excuse, even if you say, "We can't do it in Europe." Tesla is doing it in Europe right now. There's no excuse. Do you want to...?

To make a long story short: Neuralink is a smartwatch-sized device that sits on top of or inside your head, and what's it supposed to do? First, do we think it's possible and do we think it will work? That's the question. I think it's possible and I think it will work. In fact, it's already been tested that it works with a bandwidth of 124 characters per second up and down. And the idea is that it will go faster, faster, faster. So it looks like it's already worked. Do we think it's going to grow? Yes. Who else is doing this? Is Siemens doing it, too? Is Philips doing it, too? Is Pfizer doing it too? Are they as far along as Neuralink? No. No, they aren't. Who else is doing it? No one else is as far along as Neuralink. All right.

Is it going to work? Will it improve? I'm of the opinion that it will work. What will happen if it works? You won't need a smartwatch if you have one on you or in you. Will you still need a cell phone then? No. Wait a minute, what's the market cap of Apple? Huawei's? Do they make a device like that? No. Will they be able to catch up? No. So that means Tesla will probably have the entire market cap of Huawei, Apple, and Samsung. And what about a laptop? Do you need a laptop? No. And what about Internet access and connectivity? Do you need 4G, 5G?

What is Starlink? Starlink is faster internet with lower latency on every square meter of the Earth, on every single square meter of the Earth, on Mars and on the Moon. If this works... First, can it work? Yes, it's already working in sample markets. Okay, so it can work. Is it going to grow? It looks like it will. Yesterday, another 60 Starlink satellites were launched. Will it continue to grow? It sure looks like it. If it works, what does that mean for traditional telecommunications? It could shake it up.

Sohrab:
That's a good question.

Joe:
Okay. And what about mass transit? What about German rail? What about Japan Rail? What about Amtrak? What if there was a 1,000 kilometer per hour train, and it's not even a train, it's your Tesla, which is super comfortable and works in COVID times because it's your own vehicle that can go through those tunnels at 1,000 kilometers per hour? Okay, first of all, is that possible? Yes, it's already up and running for test operations in Las Vegas. Is it going to be expanded? Well, there are already bids for three more cities in the U.S., so it could be expanding. Will it expand? I don't know. I think it will. If so, what will that mean for Deutsche Bahn, Japan Rail and Amtrak? What is the combined market capitalization of all these companies? So the real takeaway is not only can we do it, but if you don't do it within 10 years, your company will die. So that's before a lot of us want to retire. And Musk keeps saying he doesn't want a monopoly. He says, "You can lease Autopilot. You can lease our technologies."

So I would say my advice is to get ahead of the curve. Try to lease one of these technologies now while your business is still strong. Try an agile transformation, and a real one, not a fake one: "We have daily scrums, but a one-year plan." See if you can really empower some employees, e.g., with budgets rather than an annual review. See if you can really let go of annual reviews and not ask people about their educational background in an interview, just ignore it. See if you can do that. See if you can really do that. Because if you can't do that, you're not going to have a job unless the government steps in and provides mass unemployment insurance. And that's it.

Please, somebody tweet me and teach me how to say this in a really loving, embracing way, because I think anybody could do this and it could be really fun. I never had as much fun at work as I did at Tesla. And yes, the hours were long, so maybe we could do something with the shifts. Maybe we could do four six-hour shifts instead and have lots of family time. And maybe we could set up places for the family and the pets, like Agile Testing Days and ALE do, right? So maybe we can really spice up the incredible.

So, let's do that, and I would say we really need to embrace that, otherwise we're going to be like newspapers or the horse and buggy industry. The agile wave is pretty big right now. And I would say let's surf that wave instead of holding on to our waterfall office that looks like it's under the agile tsunami.

Sohrab:
I think that's a good closing. Joe, thank you so much. I'm sure we'll have you back with us at least within the next 12 months.

Zuzi:
Thank you so much. That was a great story.

Joe:
I've written a book. Please read it. It's called "Scrum Master: The Business Performance Agile Training Seminar." What's special about it and other agile books is that it connects most of what we know about agile to the business success of companies like the Musk companies. So please read my book, Scrum Master. I have two agile children's books, "Everyone is Santa Claus" and "Good Night, Nut, Good Night." Please read them to your children. And the platform of Sohrab and Janet and many other people, the Scrum Academy, where I am now. Please check out my courses in the European time zone on agile hardware and Scrum Master for Business Performance, Product Owner for Business Performance, on the Agile Academy, which is built on a new platform, Agile TMS, which I'm really impressed with how Sohrab and Janet and Zuzi have contributed to it. A big thank you to all of you. Have the best day ever.

Sohrab:
Thank you, Joe.

Zuzi:
Thank you.

Sohrab:
Thank you very much, Joe, thank you very much. All right. So, to everyone.