Robots are taking over sports.
Everywhere you look, you can find them from cameras to in-stadium concessions.
And now, they’re on the court.
Introducing Vive Tennis is the world’s lightest, smallest & fastest AI-powered ball retriever.
Instead of players wasting time and effort collecting loose balls, Vive Tennis’ robot does the work for you.
So regardless of how you might feel about robots, they’re here to stay and ushering in an exciting future in the world of sports.
In this interview with CTO Javad Amirian, we dive into how the company started, his thoughts on the future of the sports tech industry, and the challenges of building a physical product company.
Hello, who are you and what is the name of your company?
My name is Javad Amirian, and I am the co-founder and CTO of Vive Tennis.
I have been into robotics and AI for the past 10 years and have a Ph.D. in Robotics from INRIA, the leader of AI research in France.
Some of my friends call me the “Father of robots”, as I spend most of my time a day developing them, babysitting them, teaching them to move, to see, to think etcetera.
The story for Vive Tennis goes back to my undergrad years and my experience in RoboCup competitions.
That time, I started forming a student robotics team, with my co-founder Sina, and a couple of other classmates, to build soccer-playing robots for Robocup competitions.
Those robots used to play with a golf ball in teams of 5 vs 5 and they play with full autonomy without any human intervention.
A few years later, a friend of mine (Behzad) who was more into tennis and entrepreneurship saw our robots and asked me if we could build a robot that could collect tennis balls using a similar technology.
What was the process of getting the company off the ground?
The proposal was very interesting to me. I had spent years working on robotic projects just for the sake of research and a bit of fun, no business, no money, and I even had to pay from my own pocket to build those robots.
But this sounded like a real-world problem and business opportunity. It’s like you love ice cream and you get an offer to eat ice cream for free and even get paid for it. Who would say no to that!?
We knew that there was already a robot in the market that could collect tennis balls in mass, which was more suitable for training sessions, where you hit tens or hundreds of balls, but we wanted to build a robot that you would use in a real game, as a real ball kid.
The idea was to make tennis smoother and more enjoyable for recreational players, by having a robot that could collect balls while the game was on.
We then devised this slogan: “Vive Tennis, Keep the rally going“.
We defined a set of 4 design principles that we wanted to follow:
1. Before anything else, the robot’s Speed should be high enough to keep up with the game.
2. The robot should be Autonomous and need as little human intervention as possible.
3. The robot should be Light and Small, and then portable for end-users.
4. The robot should be able to work on Asphalt hard courts, which are more challenging than indoor courts.
We called these principles “SALSA” and started building the prototype to see if it was feasible. At that time, I was still working on my Ph.D. and my co-founder Sina also had his full-time job doing hardware design. So we both had to spend our evenings and weekends on this project to take it off the ground.
Sometimes it was to simply validate a piece of code, to simulate a new algorithm, or to build a new prototype. Sina would take care of the hardware part, and I would take care of the software and AI part.
Behzad, who had already successfully launched a few startups, would take care of the market validation and pitching to investors.
By the beginning of 2021, we had convinced three angel investors, all friends and family, to invest in the project.
And we were ready to officially kick off Vive Tennis. We started by launching a Crowdfunding campaign on Indiegogo, very early in the project. We wanted to test the market and see if there was a real demand for the product.
At that time, the product was still far from being ready, and I believe convincing people to pre-order a product that does not exist yet is a very challenging task.
Yet, we managed to get some backers and pre-orders and a ton of feedback from people who were interested in the product.
Many of them would say they need to see the product in action before they can make a decision.
We then decided to build a full functional unit, before any further marketing or sales activities. I knew that building a market-ready product would take a full team of engineers, and might take more than a year.
But we were ready to take on the challenge, so we started growing the team and onboarding new engineers and designers. The first MVP was ready in about 6 months, but it was nothing close to what we had in mind!
Quite bulky, heavy, and as stylish as a wheelbarrow!
The ball grabber was overly complicated and the robot was very slow.
We started at the same time designing the next iteration, later we started to call it “Vive Tennis Alpha” which was supposed to be the first testable version of the product. We incorporated a “kicking” mechanism into the robot to be able to kick the balls back to the players, which was a very smart solution to save time and energy.
We also started onboarding a very top-notch industrial designer, Ernesto, who had already experience working with Nike and Home Depot.
In the next 6 months, and by the middle of 2022, we had Alpha tested on the court. We started communicating with some players and potential partners to get their feedback.
Very soon, we were invited by Head Sports to showcase our product in their court in Austria.
It was a very exciting moment for us, as we were going to showcase our product to one of the biggest players in the tennis industry. The last few weeks before our trip to Austria were very intense, as we had to make sure everything was ready for the showcase.
The demo went almost well, and they decided to run an internal survey to get feedback from their sales team. We received both positive and constructive feedback. Their sales team across different countries sent us their feedback, and it was funny to see how different people from different countries had different opinions about the product.
For example, the one from Spain would say “This might interest lazy people” or “If you play tennis, you want to do some sport”. The Japanese reaction was “This is a unique and smart product” or the Swedish one would say “It can be very useful for seniors”.
But there was a common point in all the feedback, they all wanted to see a faster robot to keep up with the game. And some were annoyed by the noise of the robot. We then came back and started working on the next iteration, which we called “Vive Tennis Beta”.
By the middle of 2023, we finished the Beta version, by increasing the speed by almost 70 percent, reducing the noise thanks to a new gear design, and a much more stylish and portable design. In the meantime, the robot’s software and autonomy were always being improved, to detect balls from 12 meters away, memorize previously detected balls, and better localization and navigation algorithms.
How is the company doing today and what does the future look like?
We are now, almost 3 years after starting this project, in the final stages of the product development. Building a physical product, especially a robot, is very challenging, time-consuming, and money-hungry.
In recent months, we have been working on preparing the first batch of 15 robots to be shipped and tested by our early adopters.
We still do not have direct sales, and these units will be delivered to some of our crowdfunding backers and some of our partners for testing.
We are talking to tennis clubs in Toronto to start a pilot project with one of them very soon. We are also in talks with Tennis Australia and we hope to bring Vive Tennis to the Australian Open in the coming years.
Initial tests have been done at Melbourne Park, and we are working on the feedback to improve the product.
Through starting the business have you learned anything particularly helpful or advantageous?
Starting a hardware product for the market has been a new challenge compared to my previous experiences with AI and robotics projects.
One key lesson I’ve learned is that building a commercial product is vastly different from working on a research project. Market-focused development is impacted heavily by external factors.
For instance, during the COVID-19 pandemic, the global chip shortage significantly affected our project. We faced tremendous difficulties finding the necessary components to build a prototype.
One of the biggest hurdles was managing communication among team members, especially since we were spread across various countries and time zones.
It was almost as challenging as getting robots to communicate effectively. Ensuring that everyone—and every robot—is on the same page is crucial for success. A major part of my role has been to keep the team aligned and to develop infrastructure for monitoring our robots’ internal states and performance.
We’ve also been fortunate to be working at a time when AI and robotics are advancing rapidly. The active open-source community has been a tremendous resource, allowing us to utilize existing tools and libraries to accelerate our development process.
Additionally, the advent of generative models like ChatGPT has revolutionized our software development, providing tools and capabilities that were unavailable just a couple of years ago.
Navigating these challenges and opportunities has taught me the importance of adaptability, clear communication, and leveraging community resources to enhance our capabilities. These lessons continue to guide our approach as we develop and refine our products.
What are some predictions you have for the sports tech industry over the next year?
I believe several tech trends will significantly influence the sports tech industry soon.
We’re currently seeing rapid advancements in AI and generative models, which are opening up new opportunities for analyzing data, enhancing player performance, and boosting fan engagement in sports. And I expect even more developments in this area.
Looking ahead, I think sports tech will become more democratized, meaning recreational players will soon have access to the same technologies that professionals use. This includes everything from line calling and game analysis to personalized training sessions.
Additionally, I see great potential in augmented and virtual reality. These technologies could bridge the gap between video games and actual sports, providing a more immersive experience.
As for my personal interest in robots, while it might still be a while before we see robots competing against humans or being regular practice partners, we are moving in that direction.
A good example is the RoboCup competition, which aims to have a team of robots capable of playing soccer against the world champions by 2050. So, the idea of a robot challenging a top tennis player like Djokovic might not be that far off.
Can you talk more about building the team and the importance of getting the culture right from the beginning?
I believe that assembling the right team is crucial for the success of any startup. Great ideas and plenty of capital are significant, but they don’t ensure success on their own.
A dedicated and skilled team, however, can make all the difference.
I’ve been fortunate to gather a group of exceptionally talented and passionate engineers and designers. For any founder, this should be a top priority—creating a team aligned with the same vision and values.
It’s the team’s capacity to collaborate effectively, communicate openly, trust one another, and embrace feedback that truly drives a startup to success.
How different or difficult is it to build a physical product in sports than building software or something digital?
I believe overall, it has become easier compared to like 10 years ago, to build a physical product, thanks to 3D printing machines and rapid prototyping which enable you to iterate much faster and cheaper than before, but still building a physical product is much harder—maybe 5 to 10 times more so—than building software.
There is no “Undo” button in the physical world. And a small mistake can cost you a lot of time and money. The stack of technologies you need to cover is much wider, from mechanical design, electronics, and software, to industrial design, and user experience.
However, that doesn’t mean you shouldn’t launch your hardware startup. If everyone focused only on software, we wouldn’t have innovations like the iPhone, Tesla, or Vive Tennis. I believe one of the next big things in tech will be robots, and I am very excited to be part of this journey.
Are you currently hiring and if so, where can people apply and find out more?
Right now, we’re not hiring as we need to finish our current product development cycle and validate it in the market. However, we are always on the lookout for talented engineers and designers to join our team in the future.
We’re also open to collaborations and partnerships.
If you’re interested in our project or think you might want to work with us later on, please don’t hesitate to reach out. You can contact us via email at hello@vivetennis.com.
