Unleashing the Speed: Exploring Software in Formula 1 with Alfonso Ferrandez - E1

In a sport where milliseconds are everything, Alfonso Ferrandez wrote one piece of software that saved not just seconds, but days…

Decision making at the pinnacle of motorsport is essential, and while F1 teams were using Excel, Alfonso’s background was the catalyst for a revolution in his team.

Alfonso shares his utterly fascinating, unique experience on the series premiere of Great Software People, taking us through the origins of the tech-accelerated racing we see today.

Episode highlights

• “I need to do lots of things, my brain needs to context switch. So I thought the best thing to do with this is actually do this for lots of different companies, so I became a fractional CTO.” - 3:30 - Alfonso Ferrandez

• “When you’re competing with Mercedes & Red Bull spending hundreds of millions more, we were looking to spend a fraction of that on tech that’ll make us leap forward.” - 10:20 - Alfonso Ferrandez

• “I wrote some software to calculate tyre degradation in F1 in a few weeks, and they were blown away. Imagine if we had a whole team to write this app…” - 13:50 - Alfonso Ferrandez

• “For our team it was transformational - we went from taking a day or more to make decisions, to literally within two laps.” - 22:20- Alfonso Ferrandez

• “You get this idea that F1 is super glamorous, with yachts and champagne and all this stuff. That’s probably true for a few of the races, but it’s incredibly hard work.” - 28:40 - Alfonso Ferrandez

LISTEN HERE:

Spotify:

Apple:

Amazon Podcast:

Rich 00:00

This is Great Software People, a podcast sponsored by HeadChannel.

So, with me today on this episode of great software people is Alfonso Fernandez, who I’ve known for quite a few years; we worked together in a couple of agencies back in the days in London. And it’s great to have you on the show, Alfonso.

Alfonso 00:24

It’s great to be here, Rich; thanks for having me.

Richard Bundock 00:29

So, I just wanted to ask you to give us an idea of your background. And we’ll talk today about how you ended up helping performance in f1. But give us a picture of a background, like how you got there; go from the beginning.

Alfonso 00:49

So, I was in academia many years ago; I came to the UK for a master’s and stayed for a PhD. While doing that PhD, I realised that the thing I enjoyed most during that period was writing the software I needed to solve my problems. Because the research needed lots of tools that didn’t exist, this was like 20-plus years ago. So you know, we didn’t have what we have now. So I had to write the code to solve the problems, which made me realise I enjoyed that part. After leaving academia to make an honest living by making money, I wanted to create software. I jumped into PlayStation games and started working for Sony because it was a perfect combination of the complex maths I’d done in my PhD and using computer science; it was like 3D and all this sort of stuff. And then from then on, I just moved on from industry to industry; I realised I enjoyed curating and building teams of brilliant people to solve complex problems, getting my hands dirty, solving those problems, but also, you know, getting people facilitating and joining the team and making sure that they were, progressing in their careers.

One thing that became apparent very quickly was that I liked the constant changes in the industry. Because from the academic years, I love the investigation and the understanding ofthings that you don’t know very well and then become an expert. And so when you change industries, say, I’m the boss in this area, in this company and technology and terms of my knowledge in this domain. It was very quickly deep diving, understanding the domain, becoming an expert to the end, and bringing people on board. And that meant two things. One is that you bring all this knowledge from other industries. And sometimes, you struggle with a new problem that you solved 5-10 years ago in another sector. Cross-pollination is very interesting because it accelerates the pace of development. After all, you bring knowledge; a perfect example was Formula 1, which we’ll talk about briefly.

So bringing all that, especially when you come into an industry that’s quite insular, you know the word that Blink is the only one to get people in from that industry. Fintechs are a typical example, right, so even in banking, they look for bankers, the base, averse to bringing in people from other industries and things, which is missing a trick, and Formula 1 was one of those as well. Basically, it kept getting bigger and bigger teams with more and more complex things. I worked for Amazon and built many global teams to solve some really cool, difficult problems. And eventually, after doing that for a while and in different industries, I realised that I enjoyed the challenge of doing all those things. But all at the same time and in a single industry. You don’t tend to do that in a single job to find that you do all those simultaneously; instead, you do them in a sequence. So you’ll do strategy and discovery and execution. But I enjoy them all, as with ADHD, I need to do lots of things as my brain needs a context switch. The best thing to do is do this for many companies. And that would allow me to do all these different bits for different companies at different stages, sizes, and challenges. And so I became a fractional CTO, which is basically where, you know, you wear the CTO hat for a lot of different companies that need different things from you, companies that may not be able to help their early startups, they don’t have the budget to afford a CTO. Still, they need the expertise to avoid making mistakes when building the platforms, or it could be some well-established company that is becoming more technologically minded. They need the advisory to ensure they’re making the right calls. They don’t have anyone in-house to help them. It’s a cool job because I get to play in many different industries, which is what I love doing and a lot of different problems. So yeah, that’s pretty much what I do.

Richard Bundock 04:40

Yeah, so basically, it was a PhD in industry, and then, you know, you worked out that you just love learning new things, and then now you’re running Indigo Labs.

Alfonso 04:48

Yeah, that’s my consultancy.

Richard Bundock 04:49

Yeah, your consultancy and as a Fractional CTO. I think we met at an agency in London.

Alfonso 05:04

Yeah, I think that was one of my first-ever contractor roles. Right after leaving my previous full-time job, I came to help in an agency, which to me was a bizarre experience because I went from a very formal background —you know, wearing a suit and tie to an office—which nowadays makes me wonder why the hell I would do that in the first place. On my first day at the agency, I remember sitting with a creative, and I didn’t know what that meant, but I was surrounded by creatives. And one of them just stood on the desk and started singing out loud. And I now realise that’s normal as creatives. So, thanks to that, it transformed how I see how the industry works and how different groups of people work within the same industry. Because I used to be always encased in the engineering and the software and things, and we’re rarely exposed to the outside of, you know, what the rest of the business did. So the agency was fantastic for that because it just, you know, you have to deal with lots of clients, but also with a quirky team of creatives. It was really, really good fun.

Richard Bundock 06:04

It was fast-paced, and no one knew what was going on. And, you know, someone had turned up on a Friday and gone, oh, this campaign’s going out on Monday, but you know, the URL in the campaign, there’s no site for it.

Alfonso 06:21

There were lots of last-minute challenges and lots of sales pitches done without knowing whether we could deliver them, basically. Then it was my job to go, okay, we will build this, and we have X days to do it. So that was quite crazy, but it was great fun. That was one of the most fun things because you have to handle lots of crazy situations with lots of really high-profile businesses. I remember pitching to a very large automotive manufacturer that had an incumbent agency for like 10-15 years, and we took it away from them because we were coming with that fresh approach of, you know, tech can do a lot for you, and this is what we are planning etc. So that was an excellent experience.

Richard Bundock 06:50

My agency experience always reminds me of an episode of ‘Whose Line Is It Anyway?’ You never quite know what’s going to happen.

Alfonso 06:52

Haha, yes, you definitely have to improvise.

Richard Bundock 06:55

You get called into a meeting and don’t know why you’re there. Yeah, we met that way, and I remember a very interesting conversation about Agile in the agency then. It was a very interesting thing. But obviously, you came out of there, and from that fast-paced bit in your first contract job, you went into other, more engineering-focused jobs, right?

Alfonso 07:46

Literally, the only reason I left that was because Amazon came knocking. And, you know, I struggled with my wife at the time because she said, “Oh, you know, the contractor world is paying so well and blah blah, all that stuff and why would you want to work there and be corporate? And I was talking to Amazon for a while. During the whole interview process, which took forever and a day, I realised, actually, internally, they were very much organised like startups and where engineering was king, right?

So, the decisions were made that each team would have to solve their problems; you have the backing of the money, but the money is like a startup, which is a great place to be, right? But it’s super demanding; there’s no getting around. But that’s when I went up.

There’s an article in the New York Times that recently came out that people were crying at the desks at Amazon and all this stuff. And I’m sure that was true in some places in our place; however, it was not like that; it was super high pressure. But you know, we work, you play hard. We had fantastic social setups, you know, the Friday evening was all like, okay, let’s switch from work to gaming or hacking or coding and stuff. So it was fantastic, you could feel the brightness of the people when you went into the building, it almost felt like it was permeating through your thinking like am I even supposed to be here because this is like a lot of really smart people and it was great fun.

But that was, yeah, prioritising engineering, invention and innovation. And then, obviously, because it’s a huge company, and you can’t afford to ruin your reputation, you have to build solid systems. That taught me about operational excellence and how you build very difficult, complex systems that actually run themselves. And if something goes wrong, it’s your job to fix it, not some other operations group somewhere else in the world, and let them worry about it. That’s not how it works at Amazon.

So it was very much you build it, you run it. And so you learn to develop it well because you don’t want to be woken up at three in the morning saying that you are losing a million dollars a minute because you’re not making sales. After all, your system has gone down or something. So we’re talking about that sort of volume, right? So it was a very different environment, but it was absolutely fantastic. Then again, I would have stayed there for quite a while. If it wasn’t for Formula One, which was the next job that came along, it was like, ‘Oh my God, I can’t believe that I’m talking to these people’. So yeah, it was quite bizarre.

Richard Bundock 10:08

So, just give us a bit more colour around how the Formula One thing happened, you know?

Alfonso 10:16

So it was a bit odd because I was talking to a friend who said, ‘Oh, you know, I know, a company that looks for really bright people and then finds a role for them. So they’ve not particularly got roles to fill. It’s just if they meet someone super interested, they will find something for them to do. Basically, I went and met the CTO the time around, who I became excellent friends with, and then he introduced me to Steven, the founder and owner of Volvo. I had a chat with him, and he had just acquired a Formula One team at the time, as you do as a hobby.

I was looking for someone I could help with technology and was thinking a little bit like the Moneyball moment, you know, where you throw tech and brains rather than money into engineering and mechanical engineering because when you’re competing with the likes of Alexa, Mercedes, and Red Bull, which were spending hundreds and hundreds of millions more than everybody else, it took a lot of work to get a competitive car.

And so he was looking for a way of what can come about if we spend a fraction of that money but spend it on technology that will make us leap forward. And I was convinced that was possible. But he was also talking to me because he was looking for aerodynamics, fluid dynamics, and expertise for the CFD. MIP is meant to be in CFD, computational fluid dynamics. And so I think at the time, I don’t know about now anymore, but when you looked at LinkedIn, I think there was only one for someone who had CTO experience and CFD.

So that was one of the blips on the radar for him. And so yeah, we had that chat, and I went away really excited about the conversation. But you know, it was just a conversation to see whether he liked me in general, as opposed to for a particular role or anything. But in the discussion, he was bringing up Formula One a lot. And I thought, well, that sounds very exciting.

And then a few days later, I got a call saying, ‘Would you like to run and be a CTO for a Formula One team?’ And so I was like, let me go to my boss and resign immediately. So I found out people in Amazon really are your friends, and people there were incredibly obsessed with Formula One. So when I told them where I was going, they’re like, ‘I hate the fact that you’re going, but I love that you get to go into sort of a thing’. So it was quite an exciting time.

And so that became the transition; I jumped off from the very advanced software engineering, and many groups globally distributed themselves to the most minor team on the f1 with a tiny development team. They called it IT, and back in 2016, when I joined, they had a head of IT that was putting a bunch of cables around the track, and there needed to be more software development. It then became apparent that it wasn’t just a small team; it needed to be more tech-savvy. I’m talking about 2016, but it would probably blow your mind with the amount of AI and stuff happening right now. But at the time, there was absolutely nothing; Excel was King. I remember going in and seeing this Excel sheet that had, I mean, I kid you not, no data in it yet. And just with the macros, it was about between 3 and 600 megabytes, just to hit an empty Excel sheet, the amount of code that is written into formulas to upload photos and send emails, literally building the whole system in Excel, which I didn’t even know you could do then.

And I thought, wow, this is obviously for a team that needs development engineers; I can see why they built a tool like this. Some of them came from the other bigger groups and said they use the same thing. So that was like, what? It seemed most of the money was going to mechanical engineering to build the car, which is what you expect in an automotive sort of motorsport.

And so I thought very quickly; I wrote a few pieces of software in my first couple of weeks. The first one was to calculate tyre degradation, which was mystifying everybody at the time because the car will go for, you know, I don’t know, in F1, you have three practices, then qualify and then race. The first three practices are basically to work out what’s the best tyre for this particular track with this type of weather, what driving style and also stuff, and it tells you how long/how quickly they degrade and how many pit stops you’ll have to do in the race for example. Hence, it helps you with the strategy.

So they’ll drive around for a few laps and get some data into the car with incredible sensors; hundreds of sensors generate 1000s of data streams. And obviously, there’s so much data that you can’t send that through telemetry. So what the car will do will send a few subsets of the data, the critical information. But when you get back to the garage, they’ll plug the umbilical, which is this fat cable that goes from the car to the roof of the garage, and it will suck out gigabytes of data. And that’s all the sensors of the calculator, and then basically at the time, they will sit there over the whole day, looking at the data and then making decisions for the day after, right, so you can see that you do an FB1. You won’t implement the changes until FB3 and list a small team because they didn’t have the resources to do it any faster.

And the first time I saw that, I thought, right, I’m going to write some software as this is ridiculous. And I wrote software that would pick up that data and give them the immediate mappings for every type of tyre and how they degraded, predicted, run lens and yellow stints, how many things we could do, how many labs, for instance, out of stuff. And they were just blown away. Like literally, that was me a couple of days coding, dragging data, doing stuff, then building an excellent user interface for it, graphs and all this sort of thing. And they were like, ‘Oh, my God, this guy’s amazing’. And imagine if we had a team to write this stuff because I’m not supposed to write like,

Richard Bundock 16:16

You didn’t you didn’t build it in Excel, though, did you?

Alfonso 16:21

No, I think it was like C sharp or something at the time. So, it was just web-based. The backend would connect to the data stores, grow the data processor, and then have nice, pretty graphs and stuff like a dashboard. And then I said to them, look, if we hire a few people to do this because I’m not supposed to be doing this, I’m supposed to be running teams, but you have no one else. And so we brought in and hired a few people, and it became, you know, we accelerated massively, and then we went from taser gradation to aerodynamic performance. And at that point, some of the sport potential sponsors realised that we were very much a technology team first. Then we started talking to exciting sponsors like Microsoft and Dell, you know, the, and rescale was, like, my favourite one because it’s like a cloud scientific service software supplier, where they provide the cloud infrastructure for, like, the big guys at like Boeing and others that build like big CAD models. Then they do stress analysis on them and all that stuff. So, for me, it was a great fit for them.

And so they threw at us lots of free computers, and I got the team that challenge and said right, guys, we need to use all these hours, computer hours, which are free, and in reality, it will cost millions of dollars, but we have them for free? What do we do with them? So, we came up with a strategy. We had a team of data, I guess, the equivalent of a data science team; we have a vehicle science team. So, the vehicle science team has a very similar approach to data science, except that they’re experts at vehicle mechanics and vehicle dynamics. And so they understand how to simulate every single bit of the car. So, like the springs, the stiffness, the angle of attack of the wing, and the performance CFD gives them, they map it out. And so you can have literally like springs and boxes model of a car that behaves exactly like the real car because it’s been mathematically modelled to be like that. And what you could do then is say, ‘Okay, well, we have a car, which is made up of this, this and this and for this particular track in Barcelona is going to be raining, it’s going to be this temperature range, this is the type of tarmac they have and discover what are the best configurations of all those elements that will deliver that best downforce, the best performance for the you know, the fast corners and all that sort of stuff.

And then what we’ll do is throw that to them to the cloud; we built a simulation engine with this vehicle science team, which will just run through, I think, towards the end, we were running about 4 million simulations before the race, and so it will give you all possible outcomes for all those combinations of all those different components. And then you will narrow it down and say, okay, well, we’re going to favour downforce, we’re going to favour whatever, and it will narrow down the configurations to the point that eventually you came up with the one suggested. We will then tell the mechanical team, who will configure the car accordingly.

So we’ve eliminated all the unknowns because now we have a mathematical simulation that tells you this is how you’re more likely to hit the performance you need without guessing it. So it takes all the guessing out, and that whole chain from having nothing to having such an advanced simulation happened within a few months; right then, we ended up with a simulated car in the strategy software. Once you decide on this component, you could drop it in the simulation in the strategy software, and it will happen in real-time or faster than in real-time. So you can say, what happens if the safety car right now or a red flag or whatever, and it will then tell you whether you have to bid or you have to do another lap and you have to change or what you have to pick for tyres, that sort of stuff because it has such an accurate model of the car.

Richard Bundock 19:55

So, you went from sort of tired aggregation to actually understanding or being able to simulate the whole car to basically simulating the whole race.

Alfonso 20:11

Yeah, so obviously, for the competitor cars, you interpret some sort of, like we knew which guys were faster. So there were parameters that you could say this car will go two-tenths faster than everybody else, like with Hamilton at the time. But then, you can place your vehicle with a very accurate simulation and drop it in the race. And you will see them going around; it was like a little scale.

So, it was still these little boxes going around the track. And then you say, okay, what happens if this one crashes and blocks the track, or if this guy triggers a safety car and stuff? It will immediately tell you what’s the best thing to do because you can make that call within seconds. And so if you’re far enough from the pit, and the call is in the pit, because there’s going to be a safety car, which means you get a cheaper, you know, a discounted version of the time that it drops to, say, takes 20-25 seconds to go through the paper and get a new set of tyres and go, the equivalent loss of time maybe like 10 seconds, if there’s a safety car because every car slows down to like six or seven times speed or whatever. And so the thing will tell you right now, go and pit because this will get you a positional advantage.

Some strategists are incredibly talented, and they have this sixth sense and tell you what’s going to happen when, but they’re super rare, so you know, technology helps with that. And now, we have moved on from that. I mean, I’m not involved anymore. But I would have been bringing in touches of AI to the whole thing, so much more complicated simulations and more realistic. But I imagine right now, if I could peer into a Red Bull or Mercedes, I’m sure they’ll have something new.

Richard Bundock 21:55

So obviously, we all think it’s the team principles, you know, making all these calls, but they’ve now got software that just runs the whole thing for them.

Alfonso 22:02

And they’re very, very clever people that run the software and configure it and put the right thing, and if you don’t put the correct data, the software is useless, right? So, it would help to have a good team of strategists, aerodynamicists, and other experts to ensure your model is suitable for the car.

Richard Bundock 22:20

So, on this strategy platform, if I can call it that, or system or programme that you put together, I mean, when running the simulations, is it just running through all the variables? What kind of mathematical models is it used to do that kind of stuff?

Alfonso 22:40

Generally, it uses game theories of Monte Carlo simulations, and you have a number of parameters. If you want to run faster than real-time, then you have to drop a few things and assume that some items are not important enough to be simulated. If you do the longer-term overnight runnings, when you do simulations, you can throw failures within 1000s of parameters to it and say, right, we will try all the combinations to see the optimal outcome. That’s what we used to do with rescale. We literally run millions of simulations.

But once you’ve got this sort of is still there, you can get this simplified mathematical model that is very, very close to their full-blown, complex vehicle. Scientists design mathematical models that can run in real-time or faster than in real-time. And so you can just throw it up there at the surface of simulation software, and it will tell you that the correlation will slow it down.

Richard Bundock 23:30

That’s fascinating. I suppose one of my key questions is, how do you know it works? So you run all of this stuff, and it tells you what it thinks might happen. But how did you know that it was good enough to rely on?

Alfonso 23:54

If you ask the engineers, it is never good enough, right? Because they just want more milliseconds out of the thing. But for our team, it was transformational because we went from taking a day or more to make decisions to taking two laps. And within FB1, we were making changes to the car that we could never have done before. So that was a testament to the fact that this enabled them to do that, which means more trial and error, which is more and more information we set effectively when you qualify. You help them as optimally as you can to get it right.

Richard Bundock 24:22

So when you made those changes and recommended stuff, you made those changes, and then the results were near the prediction. I mean, they probably weren’t bang on every time.

Alfonso 24:33

But there were improvements. So the model told you it was going to improve downforce, it would enhance downforce, but it’s unlikely that it would give you something completely different. Now, there are lots of moving parts in the actual car. The weather, the style of the driver, and the mood of the driver had an impact that you couldn’t model. And so there are no parameters, you know, when humans are involved, there’s always a bit of uncertainty.

But generally, it will say, well, if you want to maximise downforce for this particular track, these changes will help, and invariably, they did. Well, one thing that triggered an idea of mine was, I forget what race it was, but basically, we suddenly saw a massive drop in aerodynamic performance. For the car, you know the graph, you have specific numbers, and suddenly, boom, there was a step function that just dropped. And so we have to go through the footage. And we saw that in one corner that wasn’t particularly welcomed by TV and stuff, another competitor’s car hit the rear wing, and half of the rear wing, just that wingless rear wing, just flew off. Obviously, the ergonomic performance was destroyed.

And that sort of got us thinking: is there a way that we can connect the actual real-time telemetry with the model? And also with, you know, the video streams? And so you could then simulate what would happen, whether you needed to change something, obviously, a real wing you can’t change but save the front wing and if it’s safe to keep it on? Is it worth keeping it on? Are we going to get a performance degradation sufficient to justify stopping? Or can we just keep going? And that decision is generally made based on gut feeling and the strategist’s principles.

I mean, nowadays, in the current situation, you’ll get an orange or black flag or a black and white flag if your car is dangerous and needs to stop at the pits and be changed. So if you have something dangling that can fly off, you can’t stop. But at the time, when you know, if things weren’t that clear cut, and you thought, you know what I can run with this car, you could run the simulation really quickly. It will tell you whether the deficit in performance was going to make a difference, whether we will lose a positional advantage if we stop, and whether we’ll never regain it. In contrast, if we stay, we will keep that position as the duration is sufficient.

Ideas like that come in all the time because it’s such a fast-moving space. And so many weird edge cases happen all the time, but they’re not educators, really. It’s a bit like the agency. They keep throwing new things at you that you’d have to think on your feet for. So that was a great way of solving problems: thinking, saying something, and thinking on your feet to implement them.

Richard Bundock 27:11

Yeah, and so obviously, you did that work; that just started the strategy stuff, and then you began to have hints of AI. And then, for whatever reasons, you came out of that space, but they’ve carried on with that, so when you turned up, there were just Excel sheets; even the big teams were just using Excel sheets, right?

But by the time you left, all the teams started to have CTOs, teams writing software and really using the data they were getting and not relying on gut feel anymore.

Alfonso 27:57

If you’re a data-driven team, you wouldn’t be doing what we’re currently doing. The leaps in performance year on year, and the use of the fastest laps become ridiculously faster, and things that are not just mechanics, that’s also, you know, engineering and strategy. It’s funny because when I became a joiner, nobody knew who I was when I joined, right? It was weird to get a techie joining, especially a guy who didn’t come from Formula One. It’s like, ‘Who the hell is this guy?’

So I was on special projects, you know, this title, a give to everybody when they didn’t know what they were doing. So, I was the Head of Special Projects for about two months. And then, basically, when I wrote the software stuff, the management team told Steven quickly that this guy obviously knows what he’s doing, and good stuff is happening.

I think it was when I was landing in the Barcelona race and when I switched off aeroplane mode, I got the message that said company-wide, Alfonso is now our CTO, and people started looking at me because obviously, we know what CTO means, right? But in Formula One, the CTO is the chief technical officer, right? But there’s someone else that designs the car who is the Technical Director, effectively. And they were looking at me saying, How is this guy going to be, you know, CTO? And obviously, then we have to make it clear you still wanted to retain the CTO name, but now there were two CTOs. One was the car CTO to do the technology, the technical engineering guy, and then there was me, the technology guy. And so that was a bit of confusion for a while until it became clear that I wasn’t threatening this guy because I knew nothing about building cars.

So yeah, that was a bit of confusion. I mean, I’ve not been following how the teams have organised that internally. And that’s why there tends to be Head of IT and Head of Engineering, because it’s quite confusing, and you’ll get a Head of Software because the word engineering is overloaded in Formula One. And so I’m not sure what their titles are, but the concept of the CTO and we understand in the technology world, it’s now there, which it didn’t used to be, which is great to see that it’s more in position. There’s more emphasis on tech can solve all your problems; you don’t need just to throw millions out there.

Richard Bundock 30:05

So, I’ve got a bit of a weird question: you were there on race day as well, right? So you travelled with the team, went around the world, and went to all the circuits. You were there on race day?

Alfonso 30:19

I was not at all of the circuits, but I was there at most of them.

Richard Bundock 30:23

Did you have the flame retardant stuff on? Did you have the helmet on? Were you in the pit in that area? Or what was the weekend like? Like, you know, you flew there, and then what happened for the weekend?

Alfonso 30:40

I would say it is a myth; maybe it’s not a myth. But you get this idea that Formula One is super glamorous, you know, it was like, yachts and champagne and all this stuff, and that’s probably true in a few of the races. But it’s incredibly hard work. I mean, at my first race, I think it was Bahrain. And I landed there, I think we went straight to the circuit on the first day or something, and everybody, all the engineers, we have these sort of engineering suites, which are just benches with lots of computers, where all the engineers are there preparing and all that sort of stuff. And we were there until ridiculous hours, then we were there ridiculously early, you know, obviously respecting Bahrain men and stuff, but basically, I was jet lagged and like falling asleep; you have to wear the kit and all that stuff because you’re the team you have to represent the team. Then I realised these guys work 12-16 hour days, every weekend and every day of the week.

And then there are the other guys who stay to dismantle the garage, put it all in the motorhome, and take the motorhome away. So, the work is incredibly hard. I was in some European races; you have the sort of hospitality, some motorhomes you drive around Europe. And then there are the engineering motorhomes, with the engineer benches and stuff. It’s practically a big truck that you carry around. And I had a tiny, tiny room. I still remember this: the size of a phone booth with a very low ceiling, so I had to get in and sit down because I couldn’t stand up as it was at the bottom of the truck. And I had a computer, an oppressive sort of thing, so it was anything but glamorous.

But like I said, we will get out, and it will go into the pit with the garages. Then, there was a point because I was having a relationship with partners like Microsoft and all these guys, and then I became more of a CTO to show around the CTO. So they will return to the race, and then we will walk them around the garage and tell them how the comp works. They’ll see, you know, the car going out and meeting the drivers, and so it became more of a glamorous show, but that wasn’t until towards the end, which was great fun. And somewhere in between, you know, the teams they all get on fantastically in how they all work together.

There were some interesting exchanges between the teams; some provided certain beverages because they made them, and others provided other things that the others couldn’t have. And so they would meet in an alley in the paddock and exchange cans because someone was fed up with certain drinks and wanted to try other drinks. They would carry these boxes full of cables, but they were just pretending they were full of cans of drink, non-alcoholic. But it was a great mini culture development that they get on well, and they have great hospitality, beautiful food, your chef and things that kind of go with you, and they know what you like and what you eat and what you don’t eat. Having been one of the management, I could go upstairs to the VIP area of the motorhome, and they will bring you a gin and tonic as you like it and other stuff. But that was the exception; generally, it was very hard work, and everybody else worked insane hours.

Richard Bundock 34:04

It sounds like an amazing experience that you know money couldn’t buy, even though it was really hard work. You couldn’t pay for that experience, and you’ve had it. But it is absolutely brutal work.

Alfonso 34:20

Very, very and the guys on the team before the cabs came in and stuff, I remember that I think it was Barcelona or something, the principal took me for a walk down the paddock and down the pit lane, and it was literally on a Thursday. So when they were assembling, a Mercedes arrived with this massive transport, and they unloaded the car. It was covered in black, and 16 guys were rolling this car into the garage. And there was us, who sent the car parts, and the guys assembled the car on the spot on the first day, ready for track on Friday. Also, my IT guys were part of the pit crew. Also, we’re laying cable, and we had a team of 30 or 40 doing everything, and these guys had hundreds of people that would come in at different levels to the different roles. So that shows the difference between the massive teams and us.

Richard Bundock 35:14

I didn’t realise that your team was writing software one minute and then changing the tyres on the car the next.

Alfonso 35:21

The biggest guy on the team was Mike, who was more like the Head of the Team Track. He’d make sure the servers were up and running. He was super tall, like six feet or something. So he’s the guy who lifts the front of the car, right? He was the guy who had the muscle, and they’d practice when they weren’t wired and practice pit stops, which is fascinating.

Richard Bundock 35:46

I didn’t realise that I had just, you know, stupidly assumed that everyone running the pit was a highly trained specialist, and that was it.

Alfonso 35:55

They are highly trained, don’t get me wrong, but they’re highly trained on top of another job that’s announced because we’re tiny, right? I’m sure the Ferraris and the Mercedes had people who did that entirely. But right now, the Caprioli they’re all mixing roles again because you can’t have such a big team. But yeah, we were reusing people left, right, and centre; some have three roles. So it’s loading and unloading the truck, then laying cable, then the pit stops.

Richard Bundock 36:21

Oh, my lord. And then you’ve got a failed pull request, and then you’ve got to go in…

Alfonso 36:29

Haha, sorry. It’s during the rate of enrollment, you can’t touch it. Yeah, it was like that. I mean, some people—you knew that when they started, IT was not available because half of them were actually looking at the car—so it seemed quite interesting.

Richard Bundock 36:42

So you had to make everything that you wrote rock solid because otherwise, you couldn’t have something fail in the middle of the race.

Alfonso 36:55

The server racks are dismantled and reassembled on track for the next race and stuff. So, you had to ensure the hardware was fine and the software would survive reboots. So yeah, it was not your typical. I’m going to deploy this in the cloud, and it is going to run forever. I learned a lot.

Richard Bundock 37:19

It just comes to show; really, I think a lot of people have it quite easy. You know, you probably are writing software somewhere comfortable and trying to get it delivered in a slightly different environment. That’s crazy. Well, what a fantastic story about software, making a difference, and creating software in one of the most demanding environments. So, thank you so much for sharing all of that with us. I appreciate it. And, just to recap, now you’re a fractional CTO, you’re helping many different clients enjoy the difference between them and how they work. And you’re doing that under Indigo Labs, which I’m assuming you can search for online and just find them at indigolabs.co.uk. Thank you so much for being a guest on the podcast today and sharing those stories from the F1 days.

LISTEN HERE

Spotify:

Apple:

Amazon Podcast: