Shaper’s Alec Rivers on how to go from an academic prototype to a finished product
Here at RUKI, we’re conducting a series of interviews with hardware professionals who are changing the world of electronics. We want to get their opinions on what the electronics industry looks like today and how it might shape the future.
For our first series of interviews, we traveled to San Francisco and Silicon Valley, ground zero for the hardware startup boom.
In San Francisco, we spoke with Alec Rivers, a graduated of M.I.T.’s doctoral program and a co-founder of Shaper Tools, a startup building a new generation of tools for woodworking. Alec told us how he attracts former Google employees to a small company, why it is important to pay attention to design and what the future of the hardware industry might look like.
Shaper Tools is a hardware company founded by M.I.T. alumni Alec Rivers and Ilan Moyer in 2011. Their first product, Shaper Origin, allows users to create wooden objects of different sizes, ranging from a small pillbox to tables and chairs.
Shaper Origin helps DIY-enthusiasts make complex objects that can not be made with conventional tools.
How did you get interested in hardware?
My parents joked that I was born with a computer mouse in my right hand. I was always a computer kid. I was programming when I was 8, so it was really easy for me to get a computer science degree and Ph.D. However, at first I was doing computer graphics which was so far from the real world that I quickly became disappointed with my choice.
In my spare time, I loved to make things myself. I inherited my grandfather’s woodworking tools and I tried to make some projects. Soon I realized that there are a lot of hand skills that go into getting the last millimeter of detail. I could get close, but getting the last little bit was impossible for me. So I thought some technologies that I used at my day job could do that part.
And that’s how you came up with the idea for Shaper Origin?
Yes, my co-founder Ilan Moyer and I decided to create a woodworking tool that would let people do most of the work moving the tool by themselves, but use the computer to do the final little adjustments, so you can get the flexibility of a human combined with the accuracy of a computer. So we get CNC-level precision.
Are there any similar technologies or products on the market?
It’s still a really small field, but I think in ten years we will see a lot more hand-held robots. There is definitely some stuff already there. For example, I’ve seen a project called FreeD created by Amit Zoran at M.I.T. Media Lab. Moreover, similar technologies are used in medicine. One of them is the intuitive surgical system (or Da Vinci system). It’s used by surgeons. There’s a robotic arm performing surgery, and cameras on the end of the arm bring data back to the doctor, so he moves his hands and the robot arm mimics those motions exactly, but the arm performs them on a smaller scale.
How did the project develop inside M.I.T.?
I was not at M.I.T. when I had the idea and I knew that if I went back there, M.I.T. would own it. So I spent a lot developing my first prototype on my own and got intellectual property protection. Then I got back to M.I.T., polished my idea and presented it, but in such a way that I still owned all the intellectual property.
It was definitely very helpful to get back to M.I.T., my professor was extremely supportive even though I had to change the subject of my original research. I wrote a thesis on in augmented manual fabrication — building things with your hands, but letting computers help. He came up with another research idea that was related to computer vision. The project was called Sculpting by Numbers and involved 3D-sculpting with clay.
Can you tell me a little bit about the main milestones for Shaper?
As an inventor and academic type of person I thought, “Okay, I came up with the idea, now I just license it and get a lot of money.” Of course, it didn’t work out that way, because you have to actually build the damn thing. We spent a lot of time going through startup accelerators—for example, one called the MassChallenge in Boston — learning what it’s like to start a company. Then I spent about a year and a half living on my savings and working on R&D and a core technology. The prototype we did for M.I.T. barely worked. It was an academic prototype, which means it worked once then it broke. I didn’t want to get money and grow the team. Even if somebody offered me $10 million I would have said, “Leave me alone I have to keep working on this.”
When the fundamental technology worked, we raised money and started to grow the team. The most important hire was Joe Hebenstreit, who became our CEO. He was fresh from Google Glass where he took the project from an academic prototype to an actual product. That’s what we wanted to do here. Now we have nine employees and four interns.
Nearly half of our employees are ex-Googlers who worked on augmented reality glasses. They knew Joe and they were very curious to see where he is going. But I think that they were also interested because we are also working on augmenting human abilities, and Shaper Origin is not technologically very different. You have to do a lot of real-time computer vision on a really small processor. We are also coming from a lab project and trying to make thousands of products, so the same type of skills are needed.
What advice would you give to people with an academic background who are building their own hardware startups?
If you are fresh out of school, get yourself into an accelerator as fast as possible. You will never think your way into understanding how this world works, so you need to see a lot of other people going through the same process.
Running a company is nothing like academics think running a company is like. Going from a prototype to something that actually works is 10 times as much work as you need to create a prototype. Then going to something you can ship as a product is 10 times as much work again. So if you have a working prototype in a lab, you have gone only 1% of the way.
As for hiring, we realized that it’s convenient to work with new people as freelance contractors and then hire them as full-time employees. This trial period helps us to understand if we can collaborate and get on well.
Can you tell me about design at Shaper? It really stands out among other tools.
When we hired a designer, our company changed overnight. It really helped us to raise more money and to hire more people. I would echo popular advice, which is to make a designer one of your first hires. In our case it was a guy named Matty Martin who was the industrial designer at Google Glass.
Matty is the only designer I have ever worked with, and he has a very different approach. Like many others, Matty believes in Apple’s ideology where design isn’t a pretty wrap around a product, but a philosophy that you keep in mind from the beginning of a product’s development. He’s an amazing guy that just gets things done and is not a fussy designer who wants to make sure everything is perfect.
How much time have you working on Shaper? How did the industry change during that time? Which trends stand out for you?
I created the very first prototype in 2011 and the trends when I started are just getting stronger. There are easy ways of prototyping, so anyone working out of their garage can make a really professional-looking service thanks to 3D-printing and CNC-services. Things are also available cheaply, like Raspberry Pi. Now is the best time ever to start a hardware company. Even though the later stages of the development process, when you have a working prototype and actually have to produce it, are still really old-school and take a long time.
What do you think should happen to make this process easier?
There are lot of software and hardware systems that are not unified and it’s still amazing to me because I come from that pure software world where there are libraries that you can just reuse. An amazing amount of work needs to be re-done just to integrate a touchscreen into your product. You have to write Linux drivers yourself from scratch. I would love to see this [process] become more standardized because now everybody is just reinventing the wheel.
We see growing interest from investors in real-world products that have graspable impacts.
Maybe I just want to believe this, but I think people are a little bit tired of social media, apps and games. They are looking for technology that is going to change the world in the next 10 years, not for some new messaging app. I think that broadly speaking, funding is getting harder, but in our little sector it’s getting easier.
What are the technologies that are going to change the world in the near future?
I think the most disruptive technology is going to be the self-driving car. Everybody who spends 5 seconds thinking about that is amazed by what that would do.
Self-driving cars are probably a part of the reason why people are thinking about hardware right now. I suppose people might find something useful for drones to do, but I don’t know much about them. I also think there is room for around ten companies doing human robotics that are similar to our startup. There are many cases when a human and robot can work together.
To me, the fundamental enabling technology is computer vision. If the computer knows what exactly the world around it is like, you can write algorithms to drive a robot to do anything in that environment. If you think about the areas where computer vision will become tractable, it’s controlled environments like a road or a factory. In our case, we know what happens just around one particular tool, and you can get a sense of what’s going on and what needs to happen. Another big area would be factory automation on a broad level.
The Shaper target customers are makers and DIY-lovers. In your opinion, how will this movement evolve with new tools like Shaper?
Yes, makers are our beachhead market. We want this to be a tool for everybody who has a shelf in their garage and wants to make or fix something, and also professional users who are super excited and willing to spend money on unproven but exciting technology.
I think it’s going to be a huge market. And we see it now. For example, homebrewing beer is a really big thing now. Who ever thought about that 5–10 years ago? It was just not something people did. Makers are still a really niche group, but I hope that these new tools will make it a much more mainstream thing and people won’t be stunned at the idea that you can make half of the furniture in your house yourself.
RUKI is a hardware incubator, based in Shenzhen, Moscow and San Francisco.
To find out more: useruki.com