Insights from the Engineering Realm

00:00:09,465 --> 00:00:15,565
Speaker 1: My name is Matt, and I will be talking to Johannessing about leadership insights, team dynamics, and continuous learning.

00:00:17,930 --> 00:00:20,830
So we are at the point of we are talking about the education.

00:00:21,369 --> 00:00:26,705
And what interests me a lot, you have the PhD in philosophy. Right?

00:00:27,645 --> 00:00:31,699
So I'm just wondering, like, why?

00:00:32,640 --> 00:00:34,239
Speaker 2: You wanna add with my question?

00:00:34,239 --> 00:00:38,160
I got so I was working at Fraunhofer for quite a while.

00:00:38,160 --> 00:00:40,445
Fraunhofer is a research institute in Germany.

00:00:40,984 --> 00:00:48,905
It's probably best known for things like video coding and audio coding, but, like, it has all kinds of discipline, actually.

00:00:48,905 --> 00:00:55,339
So it's a really, really broad institute, probably the biggest research institute in Europe, roughly like that.

00:00:55,339 --> 00:01:03,655
And, I was in the institute for MP 3, auto coding, even though MP 3 was long long done before I joined.

00:01:04,034 --> 00:01:07,580
And, I was working there for a couple of years.

00:01:07,580 --> 00:01:10,060
It's, actually following the tariff system.

00:01:10,060 --> 00:01:13,680
So eventually, like, you only get limited contracts.

00:01:13,935 --> 00:01:15,875
Eventually you have to change your contract.

00:01:17,455 --> 00:01:20,275
And doing a PhD was actually, like, a possibility.

00:01:20,575 --> 00:01:26,320
So, my institute actually offered me to do it because they actually had interest in keeping

00:01:26,320 --> 00:01:30,399
me, and they wanted to actually offer me, like, another job also after the PhD.

00:01:30,399 --> 00:01:34,505
So that that they said, hey, don't you wanna do a PhD?

00:01:34,505 --> 00:01:42,080
And, I always wanted to do a PhD because I always, I I kind of look out for challenges, in a way.

00:01:42,860 --> 00:01:50,475
And, the only thing that actually, kept me thinking was that it's, like, financially, it's,

00:01:50,475 --> 00:01:53,375
like, even less than the tariff system in in Fraunhofer.

00:01:53,515 --> 00:01:56,875
But I I said, well, it's a chance, and I'll take it.

00:01:56,875 --> 00:01:58,940
And let's see if I can make it.

00:01:59,740 --> 00:02:04,640
And that was the thing that brought me into the PhD in the first place, I would say.

00:02:05,500 --> 00:02:12,115
And it was a challenge, and I actually didn't finish in the original time frame.

00:02:12,335 --> 00:02:18,550
I actually took it to Sweden with me, and I finished it in the evenings.

00:02:19,410 --> 00:02:25,815
And I think two and a half years or so after the original time frame, I I managed to close it as well.

00:02:25,815 --> 00:02:31,194
So, yeah, it, it was a tough tough one, but I'm happy I did it.

00:02:31,255 --> 00:02:38,970
And, also ever since I I finished it, I anyways don't really use the title, in that sense.

00:02:40,310 --> 00:02:48,075
But also in the job, the title itself or the education itself, I wouldn't say it didn't make any difference.

00:02:49,840 --> 00:02:57,605
But what did make a difference is, what I learned in terms of, like, academia work ethics, kind of. Right?

00:02:57,605 --> 00:03:00,025
So what does it mean to do research?

00:03:00,405 --> 00:03:02,105
What does it mean to write papers?

00:03:02,724 --> 00:03:06,105
It's, like, a certain diligence that you apply to your work.

00:03:06,410 --> 00:03:11,630
And, it's like a good practice in, like, technical reasoning as well.

00:03:11,770 --> 00:03:19,114
And I think I drew from from that experience quite a bit, and I still do it today, even though,

00:03:19,114 --> 00:03:25,410
like, the day to day reasoning in management and leadership is sometimes not necessarily the

00:03:25,410 --> 00:03:28,550
same as they do in a in a scientific paper,

00:03:29,090 --> 00:03:31,510
Speaker 1: but still, like, similar principle.

00:03:32,584 --> 00:03:37,864
And you said something which is not popular, especially here in Germany, to be honest. Yeah.

00:03:37,864 --> 00:03:44,319
You said, like, you have the title, you have a PhD, you don't care about so much about the title,

00:03:44,319 --> 00:03:48,495
but maybe about what you did because of this title, like the resurgence and so on.

00:03:48,655 --> 00:03:54,254
But in my experience in Germany, especially, I worked for for BMW for a while, and everybody

00:03:54,254 --> 00:03:58,790
who had a PhD, they were really like emphasizing that.

00:03:58,790 --> 00:04:01,770
And this was like a huge part of their career.

00:04:02,070 --> 00:04:04,115
So if this is what I understood.

00:04:04,175 --> 00:04:09,075
So if you want to make the career in the engineering side, you need to have the PhD.

00:04:09,135 --> 00:04:11,455
And this was what I heard here in Germany.

00:04:11,455 --> 00:04:18,080
So you are the first German guy saying this is you haven't done it for the title. It's not so, important.

00:04:18,460 --> 00:04:24,115
Speaker 2: I'm very happy that you have me here, that I can also show a different perspective on what what

00:04:24,115 --> 00:04:28,435
Germans can can do with how Germans can leave it. Yeah.

00:04:28,435 --> 00:04:32,840
It's probably like, a a special kind of ecosystems.

00:04:34,420 --> 00:04:43,325
To be frank, like, I I joined the automotive world without ever feeling like that, yeah, I'm an automotive guy.

00:04:44,345 --> 00:04:54,705
And, I I must admittedly say that I kind of despise, like, the the style that sometimes, the

00:04:54,705 --> 00:05:04,770
industry was, representing itself, like, very, very, proud, very, very, yeah, sometimes a bit arrogant, I would say.

00:05:06,509 --> 00:05:12,205
I I don't think it's necessarily still the case, and I think that I I probably exaggerated my

00:05:12,205 --> 00:05:18,205
my perspective there, back in the days, because now I know a lot of guys that actually were

00:05:18,205 --> 00:05:28,275
in the auto automotive industry, but they're everything but arrogant, so I had to I had to learn that as well. But, Yeah.

00:05:28,275 --> 00:05:32,055
I think, there are actually many people like like I described.

00:05:32,355 --> 00:05:37,015
I could totally tell from from all to the friends I had and the colleagues that I had at university.

00:05:37,409 --> 00:05:42,129
Many do it mostly for the for the challenge, I would say. Yeah.

00:05:42,129 --> 00:05:43,590
Maybe not all of them.

00:05:43,810 --> 00:05:50,795
We had a couple of folks from auto from the automotive industry, and, maybe they were following that path. Yeah.

00:05:50,795 --> 00:05:53,775
I now I know what to do.

00:05:54,449 --> 00:06:01,410
Speaker 1: Yep. And and you work in automotive industry, especially, regarding autonomous, cars for the

00:06:01,410 --> 00:06:07,665
last 6, 7 years, more or less. Right? That's a good question. I think so. Yeah.

00:06:08,525 --> 00:06:16,570
So I'm just wondering because I recently read an article about the Titan, product from Apple.

00:06:16,570 --> 00:06:19,690
So they wanted to create autonomous car, but they killed the product.

00:06:19,690 --> 00:06:21,630
So I assume it was not so.

00:06:22,995 --> 00:06:25,635
It was quite challenging to create a car.

00:06:25,635 --> 00:06:32,060
The Tesla, you have, already those system to help you with the autonomous driving and support you while driving.

00:06:32,060 --> 00:06:38,000
But I'm just wondering, like, how do you think here in Europe or in Germany when we'll see finally,

00:06:38,060 --> 00:06:40,285
like, a fully autonomous cars?

00:06:40,664 --> 00:06:44,044
Or this is still hard to hard to say?

00:06:44,585 --> 00:06:49,970
Speaker 2: It's it's definitely hard, and I had no clue when I joined, AID back in the days.

00:06:49,970 --> 00:06:54,050
I I was even surprised, like, what do you guys want from me?

00:06:54,050 --> 00:07:02,915
Like, I I don't do autonomy and, autonomous cars and, like, I'm not I don't even know, like, much about cars. I actually ate cars.

00:07:03,855 --> 00:07:10,430
So I'm, I'm a I'm a buy bicycle rider by heart, like, big time, actually.

00:07:11,610 --> 00:07:16,865
But, yeah, they didn't care about that, And I'm actually happy that they they didn't.

00:07:17,805 --> 00:07:19,085
So I just looked it up.

00:07:19,324 --> 00:07:26,720
I started 2018, at AID. So 6 years. Pretty good. Yeah. Good testament. Thank you.

00:07:27,580 --> 00:07:32,205
Just just yesterday morning yesterday evening, I wondered which year do we have now.

00:07:32,205 --> 00:07:39,164
So now you can imagine how good I am with and knowing all the dates. But, yeah.

00:07:39,164 --> 00:07:49,175
So when I when I started at AID, I I learned pretty quickly that, like, this problem that the

00:07:49,175 --> 00:07:52,235
industry is trying to tackle is pretty pretty hard. And,

00:07:54,759 --> 00:08:00,780
like, we have to use all kind of support technologies to make it happen.

00:08:01,479 --> 00:08:07,535
It's not that, well, that's what I originally thought it would be, like, that you, like, build

00:08:07,535 --> 00:08:13,860
a car, you add a couple of sensors, and then the car can see the world and then can make decisions. Right?

00:08:14,000 --> 00:08:16,580
It's, unfortunately not that easy.

00:08:17,535 --> 00:08:23,295
Even though, Tesla is trying to approach and maybe a bit like that, but there is actually a

00:08:23,295 --> 00:08:25,395
lot of more information that is coming.

00:08:25,550 --> 00:08:27,310
And, actually, also Tesla is using that.

00:08:27,310 --> 00:08:37,775
So, like, the collection of data, the creation of HD maps, like the, like this, like all of

00:08:37,775 --> 00:08:45,250
these things, they, they need to be slowly build up, and there is always this it's actually,

00:08:45,250 --> 00:08:49,810
I think, a classic nowadays, the long tail, in autonomous driving. Right?

00:08:49,810 --> 00:08:59,134
So, whatever technology you build, there might still be edge cases, that you don't fully cover.

00:08:59,470 --> 00:09:03,149
And, I think that kind of alludes or yeah.

00:09:03,149 --> 00:09:10,385
That leads to the, to the solution that we will have to wait for actual autonomous cars, how

00:09:10,385 --> 00:09:13,825
we imagine them today for a long, long time, I would say.

00:09:13,825 --> 00:09:20,300
But I could totally imagine that within certain ODDs, like certain limited, maybe sometimes

00:09:20,360 --> 00:09:25,639
fenced areas, maybe like slower speeds and so on, which is also happening, right?

00:09:25,639 --> 00:09:32,535
You can see that there, I think, in the special purpose areas, that is where, this High Court's

00:09:32,535 --> 00:09:35,275
revolution is going to, take off.

00:09:35,890 --> 00:09:43,190
And then for, like, this full self driving capability stuff where you don't need anything, the car just drives itself.

00:09:44,235 --> 00:09:49,695
Right now, I see it rather as, like, an approximate goal eventually in eternity.

00:09:51,240 --> 00:09:54,040
But, yeah, that's the that's the tough thing.

00:09:54,040 --> 00:09:56,519
So we we have to address it.

00:09:56,519 --> 00:10:04,175
And it's, it's actually not that hard to imagine because even as humans, you cannot drive everywhere, right?

00:10:04,875 --> 00:10:10,270
Go to, I don't know, Poland is maybe not that different from Germany, but go to China, and will

00:10:10,270 --> 00:10:18,590
you just sit in the car and drive safely in China as a German or Polish guy? Probably not, I guess. I wouldn't dare, actually.

00:10:18,590 --> 00:10:24,355
I was I was actually afraid to death when I first came to China and sat in a taxi.

00:10:25,295 --> 00:10:29,240
So, I wouldn't I wouldn't dare to drive there, let alone cycling.

00:10:30,660 --> 00:10:36,714
So it's it's actually a tough a tough problem, and, it has a lot of edge cases.

00:10:36,714 --> 00:10:39,455
And so I think, we we'll have to be patient.

00:10:40,690 --> 00:10:46,370
But, yeah, I think the industry also shifted a bit away from this big bang solution, right,

00:10:46,370 --> 00:10:51,585
where we just build this a robotaxi, and off you go and everything's fine. Right?

00:10:51,585 --> 00:10:57,380
Waymo is still, like, on that track, but, I think they can do it because they, first of all,

00:10:57,380 --> 00:10:58,740
they also limit their ODD.

00:10:58,740 --> 00:11:00,540
It's visible also from the videos.

00:11:00,540 --> 00:11:07,915
And, while they they just sit on, seemingly infinite amount of money. Right?

00:11:07,915 --> 00:11:14,890
So that is also good driver for building stuff, technical stuff, but not everybody has that.

00:11:14,890 --> 00:11:20,170
And, the others, they have to build intermediate products, and that means, like, driver assistance

00:11:20,170 --> 00:11:28,504
systems, and, like, the the going through the different levels 1 by 1, building something profitable

00:11:28,504 --> 00:11:34,839
that you can sell so that you can keep going building stock for the company. Yeah.

00:11:34,899 --> 00:11:39,095
And you mentioned, interesting thing that this is a lot about the data.

00:11:39,095 --> 00:11:40,214
So you're collecting a lot

00:11:40,214 --> 00:11:48,590
Speaker 1: of data, and you kind of need to interpret, those data to kind of, adapt to the external circumstances

00:11:48,970 --> 00:11:51,070
and to drive the autonomous car. But

00:11:53,850 --> 00:11:59,265
how do you see the AI and, like, current, like, large language models?

00:11:59,265 --> 00:12:01,605
Like, does it, like, somehow help?

00:12:01,745 --> 00:12:06,920
Do you see some kind of implementations that might be in case of autonomous driving?

00:12:09,460 --> 00:12:16,135
Speaker 2: So definitely, I'm super impressed by, all the things, all the progress that is happening, like,

00:12:16,135 --> 00:12:21,069
all the generative AI, products that are hitting the ground these days.

00:12:21,449 --> 00:12:28,145
I'm I'm using them all as much as I can, because I'm really, really thrilled by it.

00:12:29,585 --> 00:12:37,400
I think well, of course, large language models probably won't drive a car, but the artificial

00:12:37,400 --> 00:12:42,220
intelligence models, they are they are really closing the gap, I think, in many technologies,

00:12:42,280 --> 00:12:44,460
and they're they're also pushing the boundaries.

00:12:46,324 --> 00:12:53,660
The only thing that, we'll have to tackle there is, for things like vehicles or cars in general.

00:12:53,660 --> 00:13:02,235
Like, we're talking about, vehicles that weigh a ton, and if they smash into anything, they,

00:13:02,235 --> 00:13:04,154
like, they create a big disaster. Right?

00:13:04,154 --> 00:13:13,850
So, you have to prove that whatever they do or, like, however they behave is safe, and they don't harm anything.

00:13:13,990 --> 00:13:17,589
Like, that's that's really tough.

00:13:17,589 --> 00:13:26,185
And especially with, AI models where, like, I'm not an AI expert, but, with the hallucinations

00:13:26,645 --> 00:13:33,500
that you see with the large language models, the certain unpredictable, unpredictable behaviors

00:13:33,560 --> 00:13:35,740
that you see with some networks. Right?

00:13:36,040 --> 00:13:39,205
Like, for human perception, they are still pretty good. Right?

00:13:39,525 --> 00:13:44,825
But to actually prove that the system is safe is a very, very different story.

00:13:45,365 --> 00:13:55,095
And, it's also different in a way because, if you sell a car, the responsibility is with the driver. Right?

00:13:55,975 --> 00:14:00,714
And you cannot prove that every person on the planet is driving safely. Right?

00:14:00,855 --> 00:14:07,290
You probably cannot even prove that every person that has a driver's license, is, is driving safely. Right?

00:14:07,290 --> 00:14:12,910
But the responsibility is, anyway, is not with a driving school, I guess, but, with a driver.

00:14:13,245 --> 00:14:19,565
But when you when you sell a product, that has artificial intelligence on it, then, like, the

00:14:19,565 --> 00:14:22,699
responsibility is with the vendor, right, with the OEM.

00:14:23,240 --> 00:14:30,965
And then, yeah, like, they want to really prove that, yeah, the the product is safe, and it's,

00:14:31,365 --> 00:14:35,625
it only behaves as desired in all possible circumstances.

00:14:36,649 --> 00:14:40,029
And, that's that's a pretty tough thing.

00:14:40,329 --> 00:14:43,630
Also, like, cars, they don't see

00:14:46,255 --> 00:14:54,540
like we do. Alright? Even even the even the, the II networks, at least the ones that I I saw.

00:14:55,800 --> 00:15:02,135
So as a as a driver, when I drive my car, I I see, like, moving objects in a way.

00:15:02,135 --> 00:15:08,715
I can, like, immediately classify them as, as humans, as trucks.

00:15:09,575 --> 00:15:15,580
I can for me, often it's enough that, oh, yeah, it's a big moving object or it's a small moving object. Right?

00:15:16,520 --> 00:15:22,415
But it's not working necessarily like that with the with the lang with the with the networks.

00:15:22,795 --> 00:15:33,880
And, like, moreover, I can actually tell from the pose of the person and the way how it's walking. Is that person drunk?

00:15:34,020 --> 00:15:35,560
Is it like a child?

00:15:35,700 --> 00:15:39,545
Is it like unstable in its walking style?

00:15:40,565 --> 00:15:46,630
Where is it likely that the person will actually enter the street, or will it actually go the other way?

00:15:47,190 --> 00:15:54,790
It's like there are so many subtle information pieces of information that are attached to, these

00:15:54,790 --> 00:16:00,935
interactors that, you can actually perceive as a human, but the car cannot do that yet.

00:16:00,935 --> 00:16:03,435
So it has to actually take a couple of assumptions.

00:16:04,060 --> 00:16:10,825
And, so the actual building of that product is very different to the way how we actually drive our cars.

00:16:11,385 --> 00:16:17,405
Speaker 1: During the discussion between us and to discuss, like, a potential topics for our today's, interview,

00:16:18,500 --> 00:16:25,220
I got the, I got the impression that the infrastructure is something it's something kind of interesting for for you.

00:16:25,220 --> 00:16:30,165
But, like, if I when I hear the infrastructure and I imagine myself, the infrastructure in my

00:16:30,165 --> 00:16:37,250
mind, I have some something like completely opposite feeling that I don't see it, like, super sexy or interesting topic.

00:16:37,790 --> 00:16:42,510
Speaker 2: Yeah. Well, welcome to my world. Yeah.

00:16:42,510 --> 00:16:43,790
I, I know what you mean.

00:16:43,790 --> 00:16:48,255
And I'm super happy that we can talk about infrastructure because this is where I'm coming from

00:16:48,255 --> 00:16:51,774
or what what I've been going for the past 6 years. Oh, woah. Okay.

00:16:51,774 --> 00:16:54,675
The Elia is trying to eat your microphone. That's not good.

00:16:55,089 --> 00:16:56,290
We have a dog with us.

00:16:56,290 --> 00:16:58,529
So, we have to fight it while we're talking.

00:16:58,529 --> 00:17:02,800
It's just like an additional challenge here. So yeah.

00:17:03,904 --> 00:17:11,610
When I when I started the AD, I I mentioned it earlier in a private conversation that, I I was

00:17:11,610 --> 00:17:14,270
supposed to actually help with a sensor interfacing.

00:17:15,690 --> 00:17:22,635
But in the in the build system team itself, like, we call it engineering process, actually,

00:17:24,855 --> 00:17:28,030
there were a couple of people missing, and, I said, yeah.

00:17:28,030 --> 00:17:31,710
Well, maybe I help out a bit, but not for norms. And yeah.

00:17:31,710 --> 00:17:40,365
Well, that was my last word about that, I guess, for So I, I, stayed in the infrastructure business

00:17:40,365 --> 00:17:42,705
ever since, and I'm I'm still in there today.

00:17:44,660 --> 00:17:51,380
So for me, infrastructure is really like the whole like, all the tooling, all the ground on which everything is built.

00:17:51,380 --> 00:17:59,615
And not only the the dumb tools, but also the, like, the integration of AI models. Right?

00:17:59,755 --> 00:18:07,250
Like, if you want to, have this, inference of the network, so, like, you apply the network in

00:18:07,250 --> 00:18:11,510
your product, there is so much infrastructure that needs to happen.

00:18:11,945 --> 00:18:18,505
And the actual, like, model tweaking and model training and so on, even for the experts that

00:18:18,505 --> 00:18:25,770
work on the NL integrations, it's it's not, like, the only work that they do. Right?

00:18:25,770 --> 00:18:31,715
They don't only to do do network tweaking, but there is already there a lot of infrastructure that needs to work.

00:18:31,715 --> 00:18:38,375
Like, all the sensor data needs to be merged highly efficiently, needs to find its way to the actual network.

00:18:39,140 --> 00:18:42,340
It has to, it has to be super reliable also, the sensor.

00:18:43,460 --> 00:18:49,765
You have to also make sure that the, like, the failure modes are are captured in the sensors and so on.

00:18:49,765 --> 00:18:57,040
So there is so much infrastructure out there, that, while in my world, infrastructure is everything,

00:18:57,100 --> 00:19:04,815
but, I think in a in a product like an autonomous car or truck, like we do it here, it's easily

00:19:04,815 --> 00:19:09,375
90% in my in my opinion. So yeah.

00:19:09,375 --> 00:19:13,630
Well, maybe not as much, but it's it's a it's a it's a huge part.

00:19:14,190 --> 00:19:21,685
And I think that is something that many people don't see or actually miss, and, something that

00:19:21,685 --> 00:19:24,745
is also, as you say, it's not necessary, sexy. Right?

00:19:25,525 --> 00:19:31,510
So infrastructure sounds boring, but it's, it's extremely powerful.

00:19:31,650 --> 00:19:35,590
And, it's also exciting, and it has its own challenges.

00:19:36,985 --> 00:19:44,045
And, yeah, you can even do it with a with a PhD in, computer science.

00:19:44,200 --> 00:19:47,260
And it's still in it's still interesting, it's still exciting.

00:19:47,800 --> 00:19:52,380
Speaker 1: So what what you're saying, like, the the the really good infrastructure is important.

00:19:52,845 --> 00:19:55,825
Probably, it's super important to reach the company goals.

00:19:56,205 --> 00:19:58,145
As you said, that's like a backbone.

00:19:58,684 --> 00:20:05,250
I assume that this might be a bit difficult to explain maybe to the business stakeholders that

00:20:05,250 --> 00:20:11,112
this is such a crucial crucial part of the business, and especially here, right, like, in in

00:20:11,112 --> 00:20:11,705
a company which you which you are or previously worked. Yeah. That's that's true.

00:20:11,784 --> 00:20:14,230
Company which you which you are or previously worked. Yeah. That's that's true. I mean,

00:20:14,230 --> 00:20:16,445
Speaker 2: here in this company, I have the benefit that I already knew the people

00:20:21,630 --> 00:20:30,830
that do are in, like, the positions, that, like, in the positions that have the power to decide,

00:20:30,830 --> 00:20:37,075
like, where do we move people and who gets hold on headcount for, like, all the issues that we need to solve.

00:20:38,095 --> 00:20:43,350
In previous companies, yeah, we added We had to learn it the hard way, because the the thing

00:20:43,350 --> 00:20:45,850
with infrastructure is that it's actually invisible.

00:20:45,910 --> 00:20:49,485
You don't see it, as long as it works.

00:20:51,085 --> 00:20:57,325
And that's also one of the tough things as an infrastructure person that you have to actually

00:20:57,325 --> 00:21:01,530
show your progress very actively.

00:21:04,950 --> 00:21:10,885
And when I said we learned it the hard way, what I meant is that the build system that we have

00:21:10,885 --> 00:21:14,985
been working on, which was, like, the most critical part of the build, of the infrastructure,

00:21:16,085 --> 00:21:18,500
it was slowing down a lot. Right?

00:21:18,560 --> 00:21:23,440
And if the build system is slowing down and you have, something like a monorepo where everything

00:21:23,440 --> 00:21:29,675
is in one large repository, right, and every single developer depends on that build system to

00:21:29,675 --> 00:21:37,590
be fast, then, suddenly, like everybody in the company realises there is some problem there.

00:21:37,810 --> 00:21:39,730
And, like, can't we fix this somehow?

00:21:39,730 --> 00:21:44,675
And then, like, the obvious solution is, yeah, well, we have to actually build the tooling that

00:21:44,675 --> 00:21:50,935
supports the load that our build system and our overall ecosystem that we like, the engineering

00:21:51,075 --> 00:21:57,040
system that we build needs to handle, and that requires, like, really, really high quantity

00:21:57,100 --> 00:22:00,140
tooling, like scaling tooling as well.

00:22:00,140 --> 00:22:06,249
If you can imagine that you need to handle, like, millions of lines of codes, you need to handle,

00:22:06,249 --> 00:22:12,870
like, thousands of builds a day, like, very diverse builds as well, like, simulations in addition

00:22:12,870 --> 00:22:18,009
to sanitizers, in addition to static analysis, in addition to, like, the plane compilation,

00:22:18,215 --> 00:22:20,475
and all the different languages that we're talking.

00:22:20,775 --> 00:22:30,210
So, it's it's a highly complex system, and it it requires, like, really good engineering, and

00:22:30,210 --> 00:22:37,475
lucky for us, Google has actually paved the way for us, with a lot of tooling, and we can reuse

00:22:37,475 --> 00:22:43,929
that tooling, and learn from it, and apply these principles also to all of these things that

00:22:43,929 --> 00:22:50,429
we add to the build system and the infrastructure in general, especially the scaling one.

00:22:51,735 --> 00:22:58,855
And, yeah, as soon as the company once went through such downtimes, so we had, like, a severe

00:22:58,855 --> 00:23:05,750
situation in Argo where actually, when the build system is down and every developer stares at

00:23:05,750 --> 00:23:11,245
the wall, like, it's a huge fire, and everybody is running around and trying to solve things,

00:23:11,245 --> 00:23:13,404
and it's really, really stressful situation.

00:23:13,404 --> 00:23:16,945
And, that usually sticks with people.

00:23:18,090 --> 00:23:24,190
And then you also make sure that, you you build the right stuff, with the right quality.

00:23:24,995 --> 00:23:30,515
So, yeah, I I'm in a lucky position now that I have the people around me and the people in the

00:23:30,515 --> 00:23:39,640
decision maker situations that, positions that we we get the support that we need, actually.

00:23:40,020 --> 00:23:45,985
Speaker 1: And, talking about the complex solutions, I think, like, for complex solutions, you need to

00:23:45,985 --> 00:23:49,365
build, like, really high performing, team.

00:23:49,770 --> 00:23:54,090
So, like, I'm just wondering what is your approach to building those things?

00:23:54,090 --> 00:23:55,710
What is important for you?

00:23:56,170 --> 00:23:57,770
Speaker 2: I used to play hand bone, actually.

00:23:57,770 --> 00:23:59,705
So I'm a, like, a team sports person.

00:24:00,585 --> 00:24:05,965
And for me, it's you know the saying, like, if your if your tool is ever, everything looks like a name?

00:24:06,825 --> 00:24:10,810
So for me, every team is like a sports team. Right.

00:24:10,970 --> 00:24:18,935
I can I can recognize that not maybe not everybody agrees with me on that, but I totally see it like that?

00:24:18,935 --> 00:24:23,275
And so far, I think it actually has a really, really good, analogy.

00:24:24,935 --> 00:24:28,230
And there are a couple of things there. So,

00:24:31,810 --> 00:24:42,305
if you if you have a really, really well oiled team, it's it feels like, like a super developer to me.

00:24:42,365 --> 00:24:49,550
So it's it's almost like a person that you can interact with, and, doesn't matter what you throw at it.

00:24:50,090 --> 00:24:58,085
It always creates, like, like, a high quality, super stable, and exactly right sized outcome.

00:24:59,505 --> 00:25:05,480
And that re like, that has some resemblance to a team, I would say. Right?

00:25:05,480 --> 00:25:09,195
Like, with football or handball, what matters is the ball.

00:25:09,274 --> 00:25:13,534
How how does the ball get to the front, and how do you get it into the goal? Right?

00:25:14,154 --> 00:25:20,930
And there is so much work that is done around, like, that goal, that starts with defense, that

00:25:20,930 --> 00:25:26,914
starts with the interactions between the individuals, that starts with understanding all, the

00:25:26,914 --> 00:25:33,174
skills that everyone has in your team, and playing with those skills, and making your decisions

00:25:33,235 --> 00:25:39,110
based on those skills, and actually also pushing each other a bit, and, like, in a positive

00:25:39,110 --> 00:25:45,255
way and supporting each other, whenever there is a mistake being done, like, somebody else is helping out. Right?

00:25:45,255 --> 00:25:52,235
And, it's not about the individual, but it's about the ball, getting the ball, as close to the goal as possible.

00:25:52,950 --> 00:26:01,350
And with everybody working together, it's really hard to to actually, win against the team that

00:26:01,350 --> 00:26:03,325
is playing really well together.

00:26:04,425 --> 00:26:11,950
And, like, if if you look at that picture, then there are so many many parallels to, like, what

00:26:11,950 --> 00:26:17,330
does it look like, if if a team is high performing, and I I draw from this a lot.

00:26:17,630 --> 00:26:23,644
And so I look a lot at the interactions that the team has, how people are working with each

00:26:23,644 --> 00:26:31,950
other, also, how people how do people react when maybe a certain person is not in the room versus

00:26:32,169 --> 00:26:33,389
when it is in the room?

00:26:34,250 --> 00:26:36,330
Do we have psychological safety?

00:26:36,330 --> 00:26:43,815
Can we say what we think, and how do people actually react when we say what we think, right,

00:26:43,815 --> 00:26:52,380
and how well do we actually know each other in a way that, we can also make decisions for the other in a way.

00:26:52,760 --> 00:26:55,740
I will maybe, like, come to that, in a moment.

00:26:56,975 --> 00:27:04,735
But it's so, it's such a such a powerful, analogy that I I really come back to it all the time.

00:27:04,735 --> 00:27:12,440
And, so far, what I at least also what I heard as feedback from the teams, it it works really, really well.

00:27:12,740 --> 00:27:16,085
And, people really enjoy, like, a good teamwork.

00:27:17,345 --> 00:27:22,885
And, yeah, that's that's my basic principle how how I do it.

00:27:24,580 --> 00:27:28,120
Speaker 1: And as a leader, I'm wondering who is your partner in crime?

00:27:28,179 --> 00:27:34,335
So you work as a a guy for the engineering, like, responsible for our engineering side.

00:27:34,475 --> 00:27:40,909
So, regarding the strategy strategy, our ideas, or, like, new initiatives, like, with whom usually

00:27:40,909 --> 00:27:44,990
you work or crush your ideas, to to see if they make sense?

00:27:44,990 --> 00:27:48,769
Speaker 2: That's a good question, actually. I'm a Yeah. A partner, Caroline.

00:27:49,245 --> 00:27:54,125
I don't necessarily have, like, an individual, that is my partner in crime.

00:27:54,125 --> 00:28:05,809
I actually, I try to use, like, almost everyone, to to, bounce ideas with, to also, express

00:28:05,809 --> 00:28:15,390
my worries or my concerns, and then think about, like, how should we do things instead, and

00:28:15,390 --> 00:28:16,750
what should we do instead?

00:28:16,750 --> 00:28:22,850
And then even then, like, even if I have a clear picture, I don't necessarily execute, but I

00:28:22,965 --> 00:28:25,125
I tried out on a couple of people. Like, hey.

00:28:25,125 --> 00:28:26,005
Like, what do you think?

00:28:26,005 --> 00:28:27,945
Should we do it like this or that? Right?

00:28:29,135 --> 00:28:39,085
It's So I really try to, like, outsource it as much as possible or crowdsource it almost.

00:28:39,085 --> 00:28:45,325
So there is very there are very few things that actually don't go through the team that I, that

00:28:45,325 --> 00:28:50,700
I decide or, like, well, decision is then actually no longer really my decision. Right?

00:28:50,700 --> 00:28:53,600
If I let it go through the team, then, like, everybody decides.

00:28:54,780 --> 00:29:01,275
But I I really try to be super transparent, actually, with everything that I do, as long as

00:29:01,275 --> 00:29:02,815
it's not damaging to anyone.

00:29:03,230 --> 00:29:04,990
Of course, it's a slippery slope. Right?

00:29:04,990 --> 00:29:09,790
You can never really tell, like, that somebody doesn't feel, like, offended by it.

00:29:09,790 --> 00:29:17,185
But, I rather try to on the side of being rather more transparent than than not, and then I

00:29:17,245 --> 00:29:20,450
sometimes need to tackle the problems that I created.

00:29:21,630 --> 00:29:28,795
But overall, it's everybody around me, and everybody also has it's similar to what I said in

00:29:28,795 --> 00:29:35,870
the beginning, or or, like, what you asked me about the high performing teams. Like, everybody has skills. Right?

00:29:35,870 --> 00:29:41,650
And the more you know people, the more also you know what you can ask them and what's what their experiences are.

00:29:42,590 --> 00:29:50,095
And then you also build an intuition in terms of, what things can you discuss and what could

00:29:50,095 --> 00:29:51,315
be, like, a great outcome.

00:29:52,740 --> 00:29:57,140
Even though sometimes I just try something, I just ask a stupid question even sometimes.

00:29:57,140 --> 00:30:03,755
Even though I even though I know it's stupid, just to, like, get get people talking. Right?

00:30:03,815 --> 00:30:07,434
It's very easy that they tell me, yeah, that's completely wrong. Right?

00:30:07,575 --> 00:30:14,010
Like, your idea is shit, and I I really enjoy that if people actually open up, and and tell

00:30:14,010 --> 00:30:18,669
me that I'm wrong, especially when I knew that I was wrong, of course.

00:30:19,054 --> 00:30:21,535
So maybe the other way around is not as as pleasant.

00:30:21,535 --> 00:30:25,695
But, anyways, I have to I have to, survive this.

00:30:25,695 --> 00:30:27,934
It's, it's part of the game, I think.

00:30:28,095 --> 00:30:33,620
We have to have open exchanges, and very often, I don't see things that the team sees.

00:30:33,840 --> 00:30:41,485
And then, yeah, we we are not going into, like, analysis paralysis or so or discussing things forever.

00:30:41,485 --> 00:30:45,265
We we we use time boxes. We say, hey. We need to decide. Right?

00:30:45,750 --> 00:30:51,510
One time I even, had a situation where the team and I, we couldn't decide, and then I said,

00:30:51,510 --> 00:30:55,665
hey, let's throw a dice, right, and it was fine. Right?

00:30:55,665 --> 00:31:02,420
We just moved on because we didn't know which decision was better, so any decision was as good as the other.

00:31:03,380 --> 00:31:08,040
And, actually, yeah, it it did also turned out to be fine. Right?

00:31:08,420 --> 00:31:14,135
And, so I I try to not make a big deal out of decisions when it's not necessary.

00:31:14,835 --> 00:31:21,370
And, when we have time, And we usually have time because when we're good prepared, like, we

00:31:21,370 --> 00:31:24,990
we gather the data, and then we can also make decisions together usually.

00:31:26,125 --> 00:31:33,005
I know that a lot of people are a bit worried about, like, having big debates with the whole

00:31:33,005 --> 00:31:40,860
team and so on, But it's really also about how you, how you approach such decisions. Right?

00:31:40,860 --> 00:31:46,795
It's not that I I make a big brainstorming session with inviting 16 team members and then ask

00:31:46,795 --> 00:31:47,835
them, hey, what do you think?

00:31:47,835 --> 00:31:49,275
Like, how should we approach it?

00:31:49,275 --> 00:31:54,450
I, of course, I make my homework, I think about, like, what should we do?

00:31:55,070 --> 00:32:02,485
And then then I make proposals, and, like, then you pretty quickly learn if your proposal was good or not.

00:32:02,485 --> 00:32:06,665
Then it's because they will take it apart as quickly as possible.

00:32:07,045 --> 00:32:17,225
And if if they don't take it apart, then you at least have the, the assurance that, okay, it's not completely off.

00:32:17,304 --> 00:32:19,145
So we can actually iterate over this.

00:32:19,145 --> 00:32:27,559
And then you iterate over it as as long as you need, to get enough confidence to say, alright. Let's try this. Let's do this.

00:32:28,340 --> 00:32:32,520
Even then, it's not necessarily settled in stone. Right? You can still adjust.

00:32:33,059 --> 00:32:36,555
It's like what we always do. We do software anyways. We can adjust.

00:32:38,375 --> 00:32:42,455
But, like, this is this is my partner in crime, I think.

00:32:42,455 --> 00:32:45,600
Like, it's the it's the people around me.

00:32:46,059 --> 00:32:54,875
And, I have, like, truly good people around me, and I'm very I'm very happy to have them. And, I yeah.

00:32:54,875 --> 00:32:57,215
Actually, that's that's a good point.

00:32:58,555 --> 00:33:04,940
I once heard the the proverb that you are the average person of the 5 people that you surround yourself with.

00:33:05,240 --> 00:33:11,340
So I try to surround myself with smarter people than I have, but then they actually, like, catch up to them.

00:33:12,105 --> 00:33:19,325
And, so far, I've been pretty, successful in terms of surrounding me with me with people that are smaller than me.

00:33:19,900 --> 00:33:22,480
And, yeah, I draw from that, I would say.

00:33:23,179 --> 00:33:27,980
Speaker 1: So what you said, it brings me to to to the question that I I wanted to ask you.

00:33:27,980 --> 00:33:35,745
So you said about the crowds, crowdsourcing or outsourcing the decision making, and, like, being

00:33:35,745 --> 00:33:40,490
transparent with the people, having, like, a bright minds around, which is, like, I think it's,

00:33:40,490 --> 00:33:45,310
like, a huge lessons learned lesson learned as a leader that you don't have to know everything,

00:33:45,610 --> 00:33:51,275
that you can ask other people, you can you can have ego in your pocket, you can ask for help. Right?

00:33:51,275 --> 00:33:53,915
Because this is not, not the obvious thing.

00:33:53,915 --> 00:34:02,570
And I'm just wondering, like, about your lessons learned, like, your moments in your career as, engineering leader.

00:34:02,865 --> 00:34:10,545
Speaker 2: So I think, like, in, very, very small thing, actually, where where I felt, a lot of time that,

00:34:10,865 --> 00:34:17,670
that was a good lesson, actually, is when I thought, oh, I don't have to go, like, run this through the team.

00:34:17,890 --> 00:34:19,589
I think this is a safe bet.

00:34:19,730 --> 00:34:21,829
Let me just, like, send this out.

00:34:22,455 --> 00:34:24,475
It it always never worked.

00:34:25,655 --> 00:34:28,695
Like, always every time there was something super stupid in it.

00:34:28,695 --> 00:34:34,610
And I just thought, like, why did I acknowledge even just let someone read through it. Right? Just a review.

00:34:34,610 --> 00:34:36,770
Like, we do this all the time with code. Right?

00:34:36,770 --> 00:34:38,945
Like, and it doesn't take that long.

00:34:38,945 --> 00:34:43,585
Just show it to someone, and that's what I actually do a lot.

00:34:43,985 --> 00:34:50,970
When I, like, when I say I'm I I outsource the decision, it's it's maybe also a bit of an exaggeration, I would say.

00:34:50,970 --> 00:34:56,395
But it's more like, I try to come up with something sensible, and then I share it.

00:34:56,395 --> 00:34:57,675
And then I ask, k, folks.

00:34:57,675 --> 00:34:59,375
Do you see anything wrong with this?

00:34:59,600 --> 00:35:03,860
And then usually, I get a couple of those out. Well, I hope.

00:35:04,640 --> 00:35:06,535
And then someone says maybe, hey.

00:35:06,535 --> 00:35:07,494
What do you mean with this?

00:35:07,494 --> 00:35:10,555
This is unclear, or I don't think this is actually true.

00:35:10,615 --> 00:35:12,954
You should probably change it in that way.

00:35:13,095 --> 00:35:19,590
And, like, these things are usually done within a few minutes, and then you can just forward it.

00:35:19,590 --> 00:35:27,035
So whenever I thought, I don't need to run this through the team, I actually learned that I should have.

00:35:28,855 --> 00:35:37,730
And especially, like, decisions around, like, personal changes, in a sense when you, like, you

00:35:37,730 --> 00:35:39,270
make someone as a manager.

00:35:40,005 --> 00:35:44,885
It doesn't hurt actually to ask everyone in the team when you share your thoughts and say, hey,

00:35:44,885 --> 00:35:51,190
I actually am thinking about, like, making that person a manager. What do you think? And please be honest. Right?

00:35:51,250 --> 00:35:54,450
It's like, what happens in the 1 on 1 stays in the 1 on 1.

00:35:54,450 --> 00:35:58,390
So, and that is super important to me as well.

00:35:58,995 --> 00:36:00,615
We have those honest conversations.

00:36:00,755 --> 00:36:05,955
And usually, like, for the most part, what I heard actually that was that people say, yeah.

00:36:05,955 --> 00:36:08,430
Well, he is anyways already acting like a manager.

00:36:08,430 --> 00:36:12,050
So you can also make him make him officially the manager.

00:36:12,430 --> 00:36:20,724
And this is, like, the easiest way ever, I think, to make a manager of a team. Doesn't always work. Right?

00:36:20,724 --> 00:36:25,030
You'll like, you not always have, like, high quotes manager material in the team.

00:36:25,990 --> 00:36:32,010
It's not only that the people need to be able to do it, but they also need to want to do it. And,

00:36:35,065 --> 00:36:39,565
so it doesn't always apply, but when it does, I think it's really useful.

00:36:39,625 --> 00:36:46,230
And I always wanted to actually interview my manager before they joined.

00:36:46,609 --> 00:36:55,285
I actually never had the chance to do it, and, if my team listens now, they probably will quote

00:36:55,285 --> 00:37:03,770
me on that, but but I I think I I, I would like to, hire a manager only, also, if the team improves.

00:37:03,770 --> 00:37:10,775
But because the team, like, the team and the manager interactions are, like, what what is building

00:37:10,775 --> 00:37:13,835
the foundation of trust in a in a company.

00:37:13,895 --> 00:37:18,250
And that has to be really well oiled, in my opinion.

00:37:18,950 --> 00:37:24,815
And so I should I should do that as well, like, to even drive that through the team and not

00:37:24,815 --> 00:37:26,675
make that decision on my own.

00:37:27,615 --> 00:37:33,875
I think that is that is maybe the biggest learning even even though they were all small pieces.

00:37:34,040 --> 00:37:40,300
But adding them all together, I think they they form the foundation for for what what I'm doing mostly.

00:37:41,224 --> 00:37:49,670
What I also learned is that, and maybe that's obvious for everyone, but, I still wanted to mention it.

00:37:50,230 --> 00:37:59,450
So, like, this psychological safety thing, or being able to say whatever you want, and it's safe.

00:38:00,515 --> 00:38:04,435
It doesn't start with saying everything that you think. Right?

00:38:04,435 --> 00:38:09,160
It just, like it's it has a certain order to it.

00:38:09,160 --> 00:38:17,705
So you have to first, lay a foundation of trust and being able to say what you what you want

00:38:17,705 --> 00:38:23,625
or being able to speak your mind completely freely is only a symptom of, like, really, really,

00:38:23,625 --> 00:38:26,125
like, well established psychological safety.

00:38:27,599 --> 00:38:32,819
I must say, I sometimes remind myself of that, painfully, at times.

00:38:33,680 --> 00:38:39,724
But that's also something that, we, like, we always have to keep in mind when we communicate.

00:38:39,865 --> 00:38:46,230
We are moving, especially as managers, or leaders in general, we are moving between different

00:38:46,230 --> 00:38:53,289
teams and we have to actually be able to play the role appropriate to the situation in which we are.

00:38:53,655 --> 00:38:58,775
Especially, it's hard when you switch between different teams, when you switch between different

00:38:58,775 --> 00:39:01,115
groups of people that you have to communicate with.

00:39:01,280 --> 00:39:08,640
If you switch also between different, culture, areas that at Argo, for instance, we had a team

00:39:08,640 --> 00:39:15,135
in Palo Alto, a team in Pittsburgh, a team in in Munich, and they all had very different dynamics.

00:39:15,435 --> 00:39:23,930
And, it took a while until, you could actually speak in the same way in the different teams.

00:39:25,510 --> 00:39:30,985
So, yeah, that is also, I think, a skill that is is super, super important to me.

00:39:31,125 --> 00:39:36,470
And I think that that is something that I would recommend every every person, every leader, every manager

00:39:42,790 --> 00:39:51,130
trying to there is an audience, and, the audience should drive what you say, or at least how you say it.

00:39:52,089 --> 00:39:58,029
You still wanna get a message across, but it can be perceived very differently. Yeah.

00:39:58,694 --> 00:40:06,474
Recently, a colleague of mine, said that, they read that, the the Germans don't embellish.

00:40:07,410 --> 00:40:11,569
So in Germany, you don't meet someone and say, hey. How are you doing?

00:40:11,569 --> 00:40:14,245
And then the other person, great. How are you doing? Right?

00:40:14,405 --> 00:40:17,605
It's more like, hi, and you come to the point. Right? Yeah.

00:40:17,605 --> 00:40:22,105
Like, no, no something around, no beating around the bush. Bam.

00:40:23,890 --> 00:40:33,715
And, if, like, if you say hi and then, like, make a statement and you do that, in the US, it's perceived very, very differently. Right?

00:40:34,335 --> 00:40:36,255
Also, of course, depending on the other person.

00:40:36,255 --> 00:40:39,455
If the other person comes from Europe and the US, it's again different.

00:40:39,455 --> 00:40:48,320
But, so, like, having having some sense of, the other person in the room is, like, super super important.

00:40:49,275 --> 00:40:55,435
And it's also something that, I I I probably will never stop learning how to do

00:40:55,435 --> 00:41:05,434
Speaker 1: it. And, and the last question but not least, I'm wondering any books, resources, I don't know,

00:41:05,434 --> 00:41:12,095
conferences that were particularly, influential, on yourself, on your journey as a leader?

00:41:12,790 --> 00:41:19,589
Speaker 2: Yeah. I I read, like, all the books that I found around, agile development in, in the early

00:41:19,589 --> 00:41:23,125
areas of well, early early ages.

00:41:23,125 --> 00:41:26,164
It's, the early ages that I learned about it.

00:41:26,164 --> 00:41:32,870
So that was already far after after you were thinking that, Like, the classics there, is what

00:41:32,870 --> 00:41:36,730
I really, really loved is, coaching agile teams.

00:41:39,395 --> 00:41:45,575
It's it's one of the best books, I think, for for, actually, it's for, agile coaches.

00:41:46,460 --> 00:41:49,680
But I think it's just as applicable also to managers.

00:41:51,980 --> 00:41:55,895
Something that is very different from that.

00:41:55,895 --> 00:42:02,154
But what I also really enjoyed is especially everything around, like, a lean startup with Eric Greece.

00:42:04,500 --> 00:42:06,599
It really struck a chord with me.

00:42:06,819 --> 00:42:14,315
And, actually, the book that is, in my opinion, like, moving the needle even further in that

00:42:14,315 --> 00:42:17,855
direction is, the product development flow.

00:42:18,049 --> 00:42:18,549
Speaker 1: Yeah.

00:42:19,089 --> 00:42:27,914
Speaker 2: From Donnie Reinhardsson. The product development flow is more about, assessing, like, product development situations. Right?

00:42:27,914 --> 00:42:35,055
Making proper decision, trying to establish proper procedures, processes, practices in particular,

00:42:36,599 --> 00:42:48,954
Whereas, coaching agile teams is, is much more around how do you behave as a coach, as a manager, as a scrum master.

00:42:50,460 --> 00:42:53,760
And, actually, there is another book.

00:42:54,940 --> 00:42:59,440
It's actually, my favorite book is, Culture Code from Daniel Creul.

00:43:00,525 --> 00:43:11,630
And that is actually about teams and, how they interact and, how high performing performing teams are composed.

00:43:12,970 --> 00:43:20,895
And I always had, like, an argument with, with my previous boss, about, like, command and control.

00:43:21,115 --> 00:43:27,550
I I totally hate command and control as you can imagine maybe now after our conversation. And he said, yeah. Yeah.

00:43:27,550 --> 00:43:33,730
But there are situations where you totally need command and control. Like, with, like, firefighters. Right?

00:43:34,035 --> 00:43:35,315
You can't say, like, hey.

00:43:35,315 --> 00:43:36,595
There is a fire over there.

00:43:36,595 --> 00:43:38,835
And then everybody thinks start thinking about, yeah.

00:43:38,835 --> 00:43:42,615
But what do we do? Right? And I totally disagreed.

00:43:42,755 --> 00:43:49,470
And it took me really long to actually be able to to assess the situation and to form like,

00:43:49,470 --> 00:43:52,695
to phrase my opinion about it.

00:43:53,655 --> 00:43:56,475
Because I also saw that, yeah, he has a point. Right?

00:43:57,415 --> 00:44:03,640
It's not a good idea to just stand around and then scratch your head and then wondering, what should we do? Right?

00:44:04,040 --> 00:44:07,980
But then I I also read the the book, The Culture Cone.

00:44:08,280 --> 00:44:13,974
And, the book goes through, several team compositions.

00:44:15,315 --> 00:44:20,660
It starts with, like, kindergarteners, like, teachers in the kindergarten, how they interact.

00:44:21,280 --> 00:44:28,385
Then it went through, like, pilots that interact in a cockpit of a crashing airplane, and, then,

00:44:28,385 --> 00:44:30,305
like, a group of peaks. Alright.

00:44:30,305 --> 00:44:39,599
So very diverse, very, like, extremely different areas of of, like, teams in particular.

00:44:39,819 --> 00:44:42,705
Also, I think sports teams is also part of it.

00:44:42,785 --> 00:44:50,885
And then one of the team formations, and I was super happy about that, was actually, like, a group of soldiers.

00:44:51,430 --> 00:44:54,730
I think it was Navy SEALs or Marines. I'm not honestly sure.

00:44:55,589 --> 00:45:04,545
But, I I there is probably no no higher definition of, like, a command and control team than,

00:45:04,545 --> 00:45:06,145
like, a team of soldiers. Right?

00:45:06,145 --> 00:45:10,220
Like, this is this is my stereotype of, like, this is command and control.

00:45:10,760 --> 00:45:20,115
And then, the author described, like, how that team actually practiced and how they trained

00:45:20,115 --> 00:45:26,460
and, how they pick, I think, Simon Simic is actually using their their principles of how do

00:45:26,460 --> 00:45:27,839
you pick a new team member?

00:45:28,460 --> 00:45:32,319
How much is it about, the performance and the skills?

00:45:32,994 --> 00:45:35,555
And how much is it about the trust? Mhmm.

00:45:35,555 --> 00:45:41,395
And, like, they also have these retro sessions where they where they optimize, and they seem

00:45:41,395 --> 00:45:43,540
to be really, really rough, those retro sessions.

00:45:43,540 --> 00:45:45,000
So they're really, really honest.

00:45:45,060 --> 00:45:47,320
I don't actually wanna be there when it happened.

00:45:48,420 --> 00:45:51,800
But it like, there was nothing command and control.

00:45:52,705 --> 00:45:56,645
It was really, really, like, for me satisfying. Of course.

00:45:56,785 --> 00:46:00,780
I immediately bought a book, and I I gifted it to my my boss.

00:46:00,860 --> 00:46:02,460
IKEA, read this, read this.

00:46:02,460 --> 00:46:11,385
I don't know if you read it, in the meantime, but, they they did all through, like, training,

00:46:11,925 --> 00:46:15,125
direct feedback, and it didn't matter who gave the feedback. Right?

00:46:15,125 --> 00:46:20,060
If you're, like, a soldier and then you you do a practice session, right, you, they actually

00:46:20,060 --> 00:46:22,380
were the, I think they were Navy SEALs.

00:46:22,380 --> 00:46:24,620
They they were the the team that actually captured

00:46:24,780 --> 00:46:25,280
Speaker 1: Lullaraug.

00:46:25,420 --> 00:46:32,645
Speaker 2: Oh, well. And they practiced that situation over and over, like, with all kind of constellations.

00:46:32,945 --> 00:46:36,640
And, like, after our practice, they came together and they discussed it again.

00:46:36,960 --> 00:46:41,599
And, like, everybody had the same voice in there. So yeah.

00:46:41,599 --> 00:46:45,700
That is my example of this is how well, what I think works.

00:46:46,145 --> 00:46:50,705
And, I, so far, am working really well with it.

00:46:51,025 --> 00:46:56,860
So I think the teams like, if you if you treat them as adults, they also behave like adults.

00:46:56,860 --> 00:47:00,720
That is one of my guiding principles as well. Yeah.

00:47:01,234 --> 00:47:07,895
Speaker 1: Awesome. Thank you. I think the last sentence, it's like a good, dot at the end of at the sentence of our interview.

00:47:07,954 --> 00:47:13,059
So thanks, Johan, for, for the the thoughts, lessons learned, and

00:47:13,220 --> 00:47:16,200
Speaker 2: Thanks, Lej Sainz. Thanks for having me. It was my pleasure.

00:47:16,795 --> 00:47:19,455
Better Tech Leadership powered by Brainhub.

00:47:20,715 --> 00:47:25,935
Follow Les Schick on LinkedIn and subscribe to the Better Tech Leadership newsletter.

‍