122: Joe Justice

Listen to full episode:

Joe Krebs speaks with Joe Justice reconnect about Agile at Tesla. Even though Joe Justice is a Certified Scrum Trainer with the Scrum Alliance he explains why “Scrum is too slow for Tesla” and shares what the electric car maker is doing differently. He also explains why Scrum is still fast for the majority of organizations but that mob programming and Open Space can have a positive impact in agile teams.

Transcript:

Joe Krebs 00:20

Welcome to another episode of agile FM. Today is back to Agile FM show just as we spoke in, I looked it up may 2020. Already on this podcast. And last time this was interesting. We talked about wikispeed a lot cars. And you were sitting, I remember that vividly you were sitting in a Tesla, we were doing the recording straight from a Tesla. And today we're going to talk about Tesla. What's the irony of that after two years later? This is awesome. Welcome back.

Joe Justice 00:53

What an amazing full circle. Joe, it's my honor and privilege to be collaborating with you and your entire audience again, today. Thank you so much. In the meantime, I did work at Tesla and operate the Agile program at Tesla. And I learned a lot. And I'd be happy to update the community on what I learned while working in the company.

Joe Krebs 01:16

Yeah, so I just recently traveled to to Austin for for business reasons and took a cab back to the airport. And I kept traveling and traveling and traveling on that highway. I don't know what number that was. But I was still passing this new headquarter of Tesla just buildings in under construction right now. But I believe somebody said it's the biggest building in the country or something like that, and in square footage or something. And it's obviously under construction. But it is massive. And it's impressive. And it shows something about growth of this company. And I want to talk a little bit about what supports that growth, I want to talk a little bit about what's the agile mindset of of Tesla, this company is so on fire that sometimes it feels like it's not transparent and what's going on. At least that's my perception, right? And you're gonna debunk this because it is actually very transparent.

Joe Justice 02:10

Bizarrely about transparency, I have not ever worked in a company as transparent as Tesla, except when I ran my own company, wikispeed, wikispeed was the accounting was public, anyone could join anywhere in the world, they could simply join the company, and they were as inside as anyone else. Tesla is the closest to that I've ever seen. And that from a company with 110,000 employees and 500,000 suppliers is shocking. Most people think by the time you have more than seven employees, you probably can't afford to be transparent anymore. And Tesla has destroyed that. For example, when Elon gives what do I want to call it? I guess marching orders. When Elon says this is important. He does it through Twitter. That's how the employees see it, too. That's why it's so important that Twitter be a public record where you'll be able to say what you believe is important whether or not everyone agrees, which is Elon stance, Elon stances, Twitter should be the global Town Square where anyone can say what they think is important and not be censored. The reason why well, well, one of them is that's how Elon uses it for the business. When Elon wants to say, here's the update of the master plan, He does it through Twitter, like employees don't get it sooner than that. They look at Twitter, too. So the company is truly led in the open.

Joe Krebs 03:42

It's very different to other companies out there, right where you have to get a legal approval for going on Twitter first.

Joe Justice 03:48

Yes, yeah. Where it's you get tweets, where where other CEOs, for example, try to sound hip and cool. But it's gone through a PR department. It's gone through marketing. It was part of a quarterly release plan that was planned six months ago. And it's completely different in Elon's case. Truly, that's when the employees get the information as well.

Joe Krebs 04:16

Unbelievable, that is fantastic. So what he was or what Tesla is in the news very lately is working from home for working from home being you know, being removed, or at least like trying to be reduced to coming back to the site. Does that impact you being in Hawaii at all?

Unknown Speaker 04:34

Well, I'm no longer an employee at Tesla Oh, yeah. The mechanics under Elon statement of you need to be back in the office work from home will no longer be accepted. except in exceptional cases and you need to make the case to Elon. There are maybe two interesting topics here for for this audience.

Joe Krebs 04:58

Yes.

Joe Justice 04:59

One is order, so I'll get it out of the way first. And that is Elon saying except in exceptional circumstances, and you need to make that case to Elon. Elon has a very effective business strategy, in my opinion, called, he calls it doing your chores. And what this means is their chores, a business leader, especially the CEO, but anyone in the business has to do every day where the company suffers or even fails. And that is, for example, you yourself, viewing what are the exceptional circumstances to work from home, not delegating that to someone who might not have your same point of view right now. Elon is so deeply involved in the company that Elon does Elon's chores every day. It's these types of conversations. Okay, talk to me to Elon, why do you think you need to work from home and no one else should? Do you understand why I believe it's important to not work from home. So that's the simpler concept. Do your chores phenomenally effective? Most leaders delegate that stuff? Elon doesn't. And many, many years in this has made Elon, the very effective manager. by some measures, Elon is the most effective manager, leader, executive of all time, by some measures. Objectively, I mean, CEO of five companies simultaneously, they're all top profit, all top impact in their fields globally. This is a new phenomenon. Nonsense, possibly the Yellow Emperor of China has this occurred. So the do your chores, ethos is part of it, then the larger issue is, why not work from home? What's the issue there? Or what's the advantage to not working from home? Why is this? Why is this so important that in Elon is limited? Right? seconds on planet Earth? Elon is investing many of those seconds convincing, people are demanding that they cannot work from home. Why is this essential to the success of these Musk companies. And the core of it is Tesla makes physical products. If you're developing software for Tesla, if you're doing procurement for Tesla, if you're doing any resources for Tesla. The goal is to be within touching distance of the machines and the materials that you're supporting. Will wait what does that mean, if you're a software developer, if you're developing autopilot, the assisted driving functionality for Tesla cars. And it's also used all across the other Musk companies for other areas, right and it has artificial intelligence stacks is part of it, etc. Like auto labeler. Your goal is to be within touching distance of the machines that are etching the circuit boards for the autopilot computer, or the machines that put the autopilot computer into the cars when they're sold. Why the feedback loop of being directly in the point of production accelerates the creative process. Yeah, it also lets you physically mob if you were writing only a virtual product. If your product were virtual, like let's say you make an app for cooking recipes, yeah. And it's completely virtual. You could be remote. No, Tesla does not actually have very many really virtual products. Even the Tesla app, which I have on my phone now has a feature called Summon, where you can call the car to you physically, it will attempt to come pick you up if it needs to leak do so. And you can open the trunk honk the horn, vent the windows, you can make the car dance, plays music and wiggles the windows up and down like an ocean wave. It's actually very entertaining in my opinion. And developers are making changes to robots to factory floor automation, to procurement automation, and to the cars automation and car sensor suites. And the cars are getting 60 new parts a day. That includes electronics and circuits. So if you're writing something for the Tesla app, you think of course I could do that remote. Well, it also controls talks to the cars. And if the cars are getting 60 new parts a day, every day you are updating your code to be working well with the cars are checking at least making sure your known stable interfaces are still valid and your test cases are still valid. And you're checking information security going back and forth to the car in the app. That is way faster if you are within touching distance of the car's manufacturer. Now there's a last part of this. I've been working with them the mob mentality who advocates mob in primarily software development, but I've been using mob for hardware development and hardware engineering, okay and bringing peace to them. And they propose and have some very intriguing data to back it up. That if you have five people 4, 3, 2 people, and you have a choice, they can work alone on different pieces of work, or they can work together on one piece of work. The quality is so much higher that the overall speed is far higher, because of the reduced time in bug tracking, bug finding and bug fixing and bug deployment. And the data appears completely sound. It appears mob is dramatically faster when you accommodate time to find, fix and deploy bugs. And people report higher job satisfaction and higher pace of skills improvement. Yeah, so if you could work remote, even in a remote mob that would be next best. In person mob has an even higher rate of learning is what they're tracking. Yeah. But it isn't. Really virtual product in person mob has advantages, but virtual mob wouldn't be next best. And of course, never work alone if you truly want to be innovative quickly.

Joe Krebs 11:18

Yeah. While this is. So I had like, a few years ago, actually, it just reminded me I had Woody Zuill on my podcast, to go into the, you know, like, I don't know if he wants to be called creative, but definitely one of the leading causes of mob programming. So that's very interesting. So that is an aspect that is also happening at Tesla and is obviously in person. And that's one of the reasons why Elon Musk is promoting in person because of mob. And then the next one would be virtual mob. And that's I think I, you know, using a lot of these tools, like zoom and etc. You know, they're great for some meetings and training, etc. Right. But I, I don't know, just by evaluating this, how effective it would be for mob programming, right. Other tools come into the mix.

Unknown Speaker 12:08

Any desk seems to help a lot, any desk through any video chat system or service? Yeah, Zoom allows breakout rooms, you can dynamically re partition sub mobs. So if you are in a larger group, like I was at Tesla, often Around 50 people, sub mobs can use Breakout Room functionality at will to form task specific mobs from larger group. And that works well. It is even easier in person, you're never waiting for a tool. No one's screen isn't sharing. I mean, if you listen to an audio recording of virtual development, remote development, there is so much waiting. Yeah, of my microphones not working? Oh, you're muted. Sorry, my headset battery died. We missed the last 20 minutes. And you just don't have those issues. You have different issues, but not those issues. Yeah, and you're in person. If pace of innovation is the only thing that matters in the long run in-person mob is the only way you would choose to work currently.

Joe Krebs 13:18

So and this comes from somebody you know, who's obviously promoting in person work, who is for many, many people on this planet on the most beautiful spot on earth in Hawaii. So you know what the chances are, if you're promoting this means you're leaving the island at some point.

Joe Justice 13:36

Here is the challenge. It is easy to talk about work. And you can talk about work remotely, that's fine. You're not actually doing work. And if what I'm doing is reflecting on what I learned at Tesla and writing books, creating podcasts, delivering keynotes remote trainings, that's fine. But I'm freaking out. I guess this is a version of what people would call Island fever. Maybe this is entrepreneurial Island fever, because I haven't built anything. Yeah. And if you actually want to create something new, create something innovative, not just talk about work. You've got to be there. You've got to physically be there. So I remodeled my condo, so I can build something,. But I'm currently shopping for factory land all around the world. I was looking in Japan yesterday, I was looking in Germany the day before. I have a small piece of factory land on the mainland United States. And absolutely my goal is to get back into product development and product production. And for that you have to be co located. And in fact, if you want to be very fast, you need to live there. You need to actually sleep there. So you have to think Where's someplace that I can go not have much overhead not have much slow down and living here,

Joe Krebs 15:04

right. But it makes total sense from a. And obviously, we can only speculate what you would you would be doing with all the land, you know. And obviously land is not something you could easily get in our way so I can see why you're exploring other territories. Tesla just won't come back to that one time because what's also fascinating is to hear that you talk a lot about agile you are a certified scrum trainer, right? Is scrum a thing? Or is it more like agile a thing is, I would my guess is correct me if I'm wrong, they would be agile, but and that the organization is thinking about change and innovation rather than a specific process. But what's what's uh, what are we did the teams look like is there like a variety of things is, you know, Spotify is very transparent also how they work. And these things, just curious. Scrum is way too slow for what Tesla is doing. But scrum has a beautiful place in the world of business. So most companies around the world have annual or yearly or longer budgets. And they allocate the majority of the finances at least at a coarse level, sometimes even at a fine detailed level for a year or longer. That's normal. That's how most companies around the world allocate budgets. That means if you're going to respond to change, like a chip shortage, or a supplier change, where a new product in production, responding to customer feedback, you would need to politically advocate for re allocating capital inside the budget cycle, which is costly. You have to make presentations, you have to make reports, you have to make friends, you have to be congenial. It helps if you're handsome, you start to get politicking, which is a massive overhead most companies have about 60% of all headcount dedicated to that, yeah, to allocating politicking and shaking hands appropriately and respectfully making reports and not actually building the product. Well, it'd be phenomenally more capital efficient. If you had near 100% of staff directly improving the product like sweating, welding, programming, nearly 100% of staff, which is Tesla's goal, because of the capital efficiency involved. That means you can't take these people's time to be negotiating for budget change. So you need another mechanism for budget to change. That's very fast. That's very nimble, right? Okay. So Scrum is a awesome step towards that, what what scrum does, from my perspective, as a certified scrum trainer, is it introduces 30 day or less budgets, we call those sprints. It's a funded release, including testing, including design, a lot of companies don't do this, even though it's part they say they're doing Scrum, and don't actually have 30 day or less budget cadences, because that means disrupting their current budgeting cycle, I get that step by step, please get there. But part of the goal of Scrum is to do that, and you won't be able to move past scrum until you do. So that is the hard work some companies still have to do. It's an interesting thing. Once you've achieved 30 day or less budget allocation, and Scrum has an accountability just for that called the product owner and maximizing the value of those iterations and a list of what those are called a product backlog. And as long as the product backlog can change with feedback at any time, you have increased your agility. I'm actually predates agile in my definition, it's actually not an agile method, but it moves you towards agility. If you actually do 30 day or less budget cycles, and have a mechanism to change what the next budget cycle is at any time called the product backlog and some coalition to do it in scrum it's accountability called the product owner. Let Scrum. If you're already at 30 day or less budgets, and you're already able to produce product within 30 days, test it certified, give it to customers. That's huge. Most companies are not there. Most companies need Scrum. If you're already there, you're 30 days or less. Now the five events of Scrum, the three accountabilities of Scrum, the three outputs are artifacts of Scrum. Those are overhead now, those are now slowing you down. And that is where Tesla is. Tesla is at 60 new part introductions a day and 61 or more part deletions a day so the total part count is going down to keep serviceability becoming simpler and warehousing becoming simpler, continuous aggressive simplification that you there is no time to do so. sprint planning, much less Scaled Agile Framework which is even slower than pure Scrum, iteration planning, big room planning, there's no room for that in any of the Musk companies. Now there's room for that in a lot of companies. Yeah, if a company has an annual longer budget, I will recommend Scaled Agile Framework to help them get to quarterly releases through the train system and bigger and planning. Once they've accomplished that I'll recommend cross functional Scrum teams that manage their own dependencies inside to get to 30 day or less budgets and releases. Once we're less than 30 days, you get, you're attempting to do what the Musk companies are doing, especially if you've gotten under a week. You can sort of keep scrum useful down to a week. Less than that, it's actually just silly, you can make it work, but it's not adding value. That is where you do what the Musk companies are doing. And you actually shouldn't call it agile anymore. Because the Agile Manifesto says we're going to allocate budget, we're going to release product with customer feedback from a couple of weeks to a couple of months with a preference to the shorter timescale. So if you're past a couple of weeks, you're past agility, but I don't know what to call it. So I still call it agile at Tesla. In Tesla, there was not a name for it. There was not Yeah, so what what would it look like? Is it like Kanban-esque? Or is it like, with like, teams that just like totally self manage? building their own way of working? Is it something that results out of the mob? What is what would you see there on the ground?

Joe Justice 21:37

It is not exactly Kanban. It's like combine if you had just one column just doing column, so like a focus. But someone who is familiar with Kanban, and WIP reduction and flow theories would be feel very at home. Yeah, and so but it's simpler than Kanban. It's like combined with one column. And it's a lot like open space. So for those of us who attended the global scrum gathering Denver, which was just last week, I don't mean to make this recording sound so old three years now. But I was I actually it ended yesterday, closed yesterday. Much of that conference, as are many agile conferences was run with a technology called open space, where people propose agenda items self allocate to which agenda they want to go to. And they use the law of two feet to change sessions at any time. They can also be a butterfly or a bumblebee. And these are actually very valuable roles. As silly as they may sound, depending on what is useful to you. Tesla does not call it open space, they actually don't name it. It looks a lot like open space. Okay. So whether you're a software developer, or a product designer, or a manufacturing engineer, or a manufacturing line worker, which by the way, you are all of those things. I mean, everyone does everything. I don't know if I can really explain that without people doing it. But it's, it's true, no matter what you were hired to do, you are going to be lifting 20 kilograms every day, putting metal into robots mixing paint, you are going to be working with CAD, you're going to be working with procurement software, you're going to be working with a lot of AI software, no matter what you were hired to do. So you walk into the company, and you are in the factory, no matter what you were hired to do, you enter the factory, and it's 12 hour shifts. So in my case, I would come in at 5am. Actually, I'd often come in at 4am. But that's another part of the story. For now, let's say your shift starts at five. So you're already there. You're at 5am. And you walk up to well, everywhere in the factory are these monitors, they're usually suspended from the ceiling, but they could be coming off pylons. They're all around. But most of the time they're above you just above head height. So you almost hit your head on them. And that's important. You can see them easily. It's also available on your phone. But you don't want to have to get your phone out to check these things. You look at them all the time. And these boards did not have a name. I have since named them. And I talked about these and I teach them. I call them justice boards. I've named them after me. So they are justice boards. What these boards show you is what are all the funded projects like what is Tesla spending money on or SpaceX? When I visited SpaceX, it was the same what is this company funding money on right now this second, these are real time they're driven by Tesla's vision. So Tesla vision is an AI stack it is not going to solve all your problems for you. But what it's very good at doing is auto labeling. looking at pictures or now video multi channel 8k video now but looking at vision and labeling what's there who Is there this employee? Okay, what robots are there? How much does it probably cost to run those robots? About how many cubic meters of the factory? Does it look like they're using? Okay, well, what's the cost to Tesla to consume about that many cubic meters of factory, and they automatically gas and the gas isn't perfect, but the value is no engineer has to think about it. It's automatic. They guessed the cost of whatever group is working. And then that group has to say, through an app on their phone, what am I doing? And how do I think we could measure the value, so they could say, We're re-programming three of the paint robots, we think we can reduce the time a car spends in paint by 11 seconds, what we're really they don't take time guessing a number, they just say we think we could reduce the time a car spends in paint, which happens to measured in seconds. Well, what the AI then does, is it calculates the value to the company, if the cars spent any less time in paint, without defect rate going up with quality increasing, and without consuming more cubic meters of the factory or more robots, or if you do, it's offset by the time reduced in pain. And that becomes a row on the Justice board. Now what I added, since leaving, I've had a chance to reflect right, so what you can do remote well is reflect you can sit and think is these boards were unique per area, and you could toggle through them on your phone. What I added is to the justice board, is there's a row for each of these. And they dynamically sort up and down based on value ratio, which is more valuable versus expense that's at the top, which is lower value versus expense. Now, when I was in Tesla, we were encouraged to guess at that, and use the law of two feet to go to the highest value product that we thought we could meaningfully add value to or we were really passionate to learn. And we thought it was worth the investment in our learning. And we were asked to make that call, there was no manager allocating yourself allocate like attending an open space conference. And you can use the law of two feet to go to a different work group anytime. And what these work groups are changes in real time because people can propose them from their phones, they can just go to a new area and start working on a new idea and propose it with their phone. Well, now the Justice board dynamically ranks highest value top lowest value bottom. And if what you propose is the bottom, and it doesn't look and you don't have a great idea of how to dramatically increase its value, you should probably use the love to feet and abandon it immediately.Right? You're just managing money.

Joe Krebs 27:44

What a great concept is like, you know, like open space. I've used it, you know, obviously in a lot of conferences and I had Harrison on here one time on on agile FM to speak what an interesting concept, right? But now to use it actually, it turns out so if I would I'm in the development stages, right? So what's super fascinating is like you're talking about about group work, you're talking about crowdsourcing of ideas. I heard recently, I forgot the reference I have to look it up with that is obviously when I heard the sentence somebody saying the next 10 years, we see more innovation that in the last 100 years combined. And, and companies obviously need to be prepared for that. Do you see that as a recipe for it? Is this because you said 110,000 employees, the company obviously grew very, very fast to that number you guys are doing or have been doing things in during your employment, you observe things of crowdsourcing? Do you think that's scalable bigger, if that company is growing? Like what's what's your advice for companies that are possibly looking at that stat saying like in the next 10 years, there's so much innovation, I'll be ready for this. What should we do?

Joe Justice 28:57

There are two enablers that I'm aware of now, one is Musk's philosophy of do your chores. And you have to do those every day. That's what prevents the company from becoming siloed. Hierarchical. Like when you join Tesla. You're given the anti handbook handbook. It's four pages long, and it's been leaked online. And it looked exactly like the one I was given. So I believe that page is completely correct. It's four pages, that is your training. And then safety training and you're in Tesla. It takes four hours total you're in, you get your work shoes, you get two pairs of safety shoes, and you get the AI apps downloaded on your phone which replaced management. So there's effectively no managers in the Musk companies. AI has replaced that which is the idea of digital self management, which moves towards the goal of 100% of staff directly improving the product, which is higher financial efficiency, higher capital efficiency, part of the anti handbook handbook, and it's very few words for agents who will be part of it. So As you can talk to anyone, you can go anywhere, you can do anything. And in fact, if you see a bottleneck, if you see a problem, and you don't go work it, we will hold you accountable for that. So it is your job no matter where you saw it. And it says you can talk to anyone, anywhere, you can talk to Elon, if talking to Elon is the fastest way to resolve this bottleneck, you must do that. And if anyone tries to stop you, they will be fired immediately. This is what prevents the hierarchy. Now, what that means, conversely, is if someone prevents you from talking to Elon, for example, about removing a bottleneck, not because you're like, hey, what kind of pizza do you like, but because you you want to I mean, you could do that, too, he lines up pretty friendly person, but especially about a bottleneck to the organization. Someone actually has to fire them immediately. And so Elon actually has to check. Is that happening? Are these words still what is running the company? So that's part of the do your chores, if you are in the position of deciding who gets paid, who doesn't who's on staff who's not who has access inside the company, who doesn't what we usually call employment. But now in the Wikipedia global contribution model, it's more like who has author access, right? But whatever, if you're in the position of determining that you have to do your chores every day. Now, the second part of that is you can automate a lot of those chores I just mentioned, digital self management, right? This is truly, in my opinion, the musk company's superpower. What Elon is doing is intensely funding, and joining the mobs to automate away wait states. If someone is waiting to ask a manager or an expert Did I do a good job? If we can, that should be an app on your phone and running on the monitors above you. So you never have to wait to talk to another human. The idea of Elon's chores are being automated to the extent they can be. And that is part of the goal is that is what x.com is supposed to be supposed to become eventually, which is a domain of just a Url domain on the internet that Elon owns an aspiration for x.com, which was tweeted to Elon, a couple of years ago, and Elon replied, Yes, that's exactly what I should do. You're right, is for it to become the parent organization that keeps the other companies lean and valuable, and radically innovating. So you don't have the founders departure complex, where when the founder when the founder departs the company dramatically slows down, it falls into corruption falls into debacle and decay. Well, the attempt to solve for the founders dilemma because Elon says, Look, by the time someone is 70, they probably shouldn't have a job, their brain is corroded, biologically, that Intel neural link does what neural link supposed to do. You should not be running a company and Elon says no one should be CEO forever. So you can count that down. That's 20 something else? Yeah, 21 years from now. Until Musk says, you know, I would be incompetent it'd be criminal derived from Elon to have Elon having financial authority of the companies on a day to day basis. So that's the timeline where x.com needs to have replaced most of Elon's chores, if not all with automation. Then you have a rules engine, like playing a video game to keep the rate of innovation high to keep the politics out to keep silos out to keep a flat nimble organization at any size. Elon, because of an aggressive workday, is able to do that across five companies simultaneously, all around the world. So you can a human can you just have to make it your passion. Just like if you love to watch sports every day. And you watched two full sports games everyday if you were deeply passionate, and on the weekends, maybe watch four. Well, if your business is that level of passion, even doing your chores, the unfun part of it, then you can scale this you can keep this innovation high so we can continue to have founders and even not founders, even board members who are at this level of performance. However, what Musk is not currently selling, not now, is any of the digital self management. So if you want to compete you better be building your own

Joe Krebs 34:50

This is awesome. Yeah. Joe this is maybe there's somebody listening right now as it's like wow, this is a very open and transparent conversation where and you know in And I want to say thank you for being so transparent. What's interesting is, I have seen some, I think it was somewhere on social media, I have seen you post something, and Elon Musk actually liked it, right? So you got a like, so I hope that people when they find this episode that they also gonna like it right? And possibly vote, vote this episode up. And because they enjoyed the conversation, for the parents out there listening to this, I decided to get my kids prepared for the workforce, and the workplace in the future. What I'm learning is do your chores and work in groups, right, and be able to work in groups? I think that's what I'm hearing when that should be promoted. What you don't want to tell your kids is that hierarchies don't matter. At least until the 18th. Right.

Joe Justice 35:51

Joe uour was hilarious.

Joe Krebs 35:54

Yeah, this was an awesome conversation, right? I love it. And looking back two years ago, coming back to where we started. Two years ago, you did not work for Tesla at that point, and what kind of vision and everything now we talked about Tesla tremendous amount of information, and that is fantastic to digest and getting some insights. Now with your little cliffhanger on I'm trying to buy land, we can only assume what's going to happen in two years from now. When we speak again. I hope it's not going to take two years to talk about again, but hey, let's see what's what's in the making always curious.

Joe Justice 36:33

Joe, it's my honor and privilege to be part of Agile FM, Agile.FM to the moon.

Joe Krebs 36:39

Thank you. Thank you for listening to Agile FM, the radio for the Agile community. I'm your host Joe Krebs. If you're interested in more programming and additional podcasts, please go to www agile.fm. Talk to you soon.