About this talk
In this talk, the speaker, an IT professional and amateur racing driver, explores the concepts of speed and velocity in the context of startups, enterprises, and Formula 1 teams. He highlights the importance of direction in achieving results, explaining how startups are nimble and quick to correct their course while pursuing goals. The speaker compares the velocity of startups and enterprises, noting that successful enterprises have mastered making everyone move in the same direction, leveraging processes that may be unpopular but necessary to avoid losing focus. He emphasizes that Formula 1 teams, despite their size, maintain a startup-like speed by integrating rapid development cycles and specialized processes. Ultimately, he argues that company size is less important than teamwork and effective processes for achieving timely results.
Full transcript
Warm up your hands. My name deer. I'm a professional IT guy and an amateur racing driver and we'll be talking about speed today and velocity. So, quickly with show of hands, who of you here has worked or works in a startup? A few people. Okay, good. We have people from startups. And who of you has worked or works for an enterprise? I would imagine much bigger number.
Okay. There were people that didn't show hands. Uh my next question is who here has worked or works for a Formula 1 team? It must be some of you guys. Uh but then good. uh you will be keeping me honest uh because our topic uh is to uh compare velocity and uh why you should listen uh to to this to this talk. Uh reason number one is
Rado uh because uh he is from the organizers and he decided that it will be useful for you uh to see how motorsports can help software. And uh number two, I spent uh years um in all the three uh options. So I got some experience and I can compare for you and share my experience with you. Uh and this is an old problem actually uh who is
obviously faster. This is a paradox which is more than 2,400 years old. Uh it is in the math books uh in the computer science 101 etc. Uh and uh if you ask a junior today what is velocity, they will most of the time tell you about speed which is incomplete because what is the difference between speed and velocity? Anyone? >> Sorry. >> Direction. Exactly. Direction is the
key difference between speed and velocity. and direction is critical to achieving results as well. So yeah uh what is the measurement that that we are using uh in software to measure velocity anybody you don't do scrum that's that's my conclusion story points exactly so uh so the story points are the units of measurement uh in software development for some and they are actually a glorified uh percent
or man months And this is how uh the whole fun and games of uh estimating and planning etc. uh uh how they start. Uh so how fast is a startup now that that we have the uh mathematical background for that. So the theoretical maximum velocity of a uh is a function of the number of people in it and the number of hours per day and the number
of days per week. you can't exceed that even with with AI you still have an upper boundary uh for this velocity and the variable there uh is is but why startups are fast because they are very fast at uh chasing one particular goal and they are very fast at course correcting whenever they detect that they are going off track with respect to achieving the goal. And uh
in this respect they are much rather nimble than fast. They are like a go-kart very fun to drive. You feel the speed. All the champions are built through this experience through go-karts through uh through startups like Nicole for example here. Uh, and there's a limit of how fast it can be. And as startups grow, there's one big thing that can hamper their speed, that can kill literally
their speed, and this is the Brownian motion, which is the complete loss of direction and complete loss of results. And there are many vectors that that can push a company, a startup, a scale up, any company uh into Brownian motion. They're listed here. So the question is if you want to win bigger races, you you will need a bigger team. What is the magic? How the enterprises
can potentially achieve uh bigger velocity as they have more people? Well, every enterprise they were um a startup once. So, so the formula is similar just the the number of people is bigger and they manage uh to win bigger races some of them. So, the the thing is the secret source is they mastered the art of making everybody move in the same direction. the ones that are
successful, they managed to to master this art. And everybody will talk uh to you about culture, about structure, about all these sort of things. But today I want to talk uh to you about the thing that everybody hates. And this is the process. And why people hate process? Because and why I hate process because stops me from doing what I want to do right now at the
moment and asks me to do what I need to do. And the best part is it does this for all my fancy uh colleagues as well. So we all hate it and we can all follow it to to go in the in the same direction. And this is saves us uh from the brownian motion. all of you who are in an enterprise and who are looking from
the sides at the at the enterprise, why doesn't it it feel that the the enterprise is actually fast? Well, have you been on an airplane? Probably you have. Uh and uh what is an airplane? This is basically hundreds of people moving very fast in the same direction and you can even sleep on the plane and still move fast uh in this direction. This is actually what happens
in a lot of uh of of the corporates. Uh so yeah that that's one of the explanations. Uh but the more relatable one uh uh for us as an audience uh is the comparison between a really top end uh CPU with top- end cores uh and a very basic entrylevel gamer GPU. And as you can see, at three times less frequency for the course, the the gamer's
GPU with a much more much simpler course can achieve 30 times the teraflops of the CPU. So there you go. So how fast is a Formula 1 team? This is this is what we started with. So every Formula 1 team has roughly 100 software developers give or take and uh what they need to do which is part of their competiti competitive advantage compared to the others is
they need to develop several critical and very proprietary software systems uh for them and they are releasing those systems every week. So the the the development cycle is literally weekly and these systems let's take one of them for example which which gives an answer to a very simple question the strategy uh system which works real time at the formula 1 race. Its purpose is to answer a
very simple question when the car needs to stop in the pits in order to gain an advantage. And to do this, this system integrates tens of billions of data points from all the other systems in order to produce hopefully the correct answer. So this is and and books they cost literally millions uh in this environment. Uh so this is how uh the the formula one teams work
and how do they achieve this? Uh they have the nimleness and the focus of of of a startup. They have the size which is bigger than a than a typical startup more like an enterprise and if you consider the entire team it is literally an enterprise. So how they achieve the speed they put on top of the the process the enterprise process and the enterprise specialization. So
there it goes. So it is mission possible uh to do it week in week out for the entire season and the next season. So what are the takeaways? Company size doesn't matter. Play the team game. Don't crash into your teammate and follow the process. And based on that, what would be the northstar company, the best company that that there can be in the software? This will be
a company promotes and grows in autonomy and personal uh effectiveness, but it also focuses a lot on quality and it is implementing strict processes for everybody to minimize the Brownian and it goes for timely results. And the last thing, it doesn't come from me. All the other smart things were mostly my thoughts, not AI. Uh but the last thing comes from one of the geniuses in Formula
1 uh from Adrien Yui. Basically, this guy uh says that adaptability is the key thing and whenever reality contradicts the numbers that you have, trust the reality, which is a very smart advice with which I want to to leave you with. Wish you make great software. Lights out and away we go. Thank you.
