Episode 12 -
Building a Startup in Educational Robotics

The following is a transcript of an episode of The K12 Engineering Education Podcast. More transcripts for other episodes are linked from the podcast main page, k12engineering.net. Extra information about the episode, including links to relevant resources, are listed in the show notes, which can be found on iTunes, SoundCloud, or your podcast player.

Transcript of:

The K12 Engineering Education Podcast


Building a Startup in Educational Robotics

Release Date:



[Pius Wong] This is The K12 Engineering Education Podcast for September 19, 2016.

[opening music]

[Pius]  I’m Pius Wong, your host today.  I’m in Austin, speaking with Connie Hu in San Francisco.  Connie is the CEO and Cofounder of Arcbotics, an educational robotics company.  Previously she advised top global nonprofit leaders on strategic planning and K-12 STEM education at the Bridgespan Group.

[Pius]  So Connie, welcome.  Thank you for speaking with me over Skype today.

[Connie Hu]  Thank you so much for having me, Pius.  This is great.

[Pius]  So for those who don’t know what your company is, what Arcbotics is, how would you describe it?

[Connie]  I would say that the mission of our company is to create robots that are easy to use, that are affordable and open-source.  We want to make robotics learning as easy as possible, and so we have a line of kits, as well as accessories, to that end.

[Pius]  Are those kits intended for a particular type of customer?  For teachers, for kids, for parents? 

[Connie]  Yeah, so the end user really -- we’re going for STEM educators, so fourth grade through college level.  After school programs.  We’re also teaching STEM, a variety of educators there.  Parents who want to prepare their children.  And also we’re getting a lot of adults who are wanting to learn robotics and programming for the first time, but, you know, maybe missed that train a decade ago, because there weren’t a lot of good resources out there.  They’re teaching themselves  with their robots, as well.

[Pius]  Oh wow, so a lot of hobbyists.

[Connie]  Yeah, definitely.

[Pius]  Would you say that your most popular products are those kits, then, and not necessarily the – it sounds like you also teach or you have services.

[Connie]  So they come very much together.  Our two main robot kits are Hexy the Hexapod and Sparki the easy robot, and they’re the two most popular by far, especially Sparki, for schools.

[Pius]  Why is that?

[Connie]  Lots of reasons.

[Pius]  [laughs]

[Connie]  For Sparki we hear from our teachers the main reason why they purchased was all the features, all the sensors.  Because a lot of time they might pay for something, and it sounds really cool in the beginning, but frankly it comes with one to two sensors, so then you run out of things to do after an hour with your students.  And so they look at ours, and it comes with, you know, over twenty-five sensors and actuators, a lot of different features, and also we provide a hundred free lessons going from beginner to advanced across the spectrum of topics.  So they feel, OK, I don’t have to invest a lot of time to create my own curriculum.  You already give it to me.  And it comes with a lot different software environments tailored to the different ages.  So, a fourth grade teacher, a sixth grade teacher, might start with our drag-and-drop programming software.  It’s a lot easier for younger students to get into.  And then more advanced levels – whether they themselves progress – our middle and high school students might to the Arduino software.  Also we just launched a partnership with Codebender to do Chromebooks, too, because a lot of schools now are asking about Chromebook software.  I think it’s just that whole ecosystem, and we have active forums for teachers and other users posting their own projects all the time.   We have this fantastic YouTube playlist of about fifty videos, all from our community.  All the schools, all the parents who’ve posted up their projects, it’s really cool to see that.  So I think they’re excited about that whole ecosystem.

[Pius]  So it sounds like you have a pretty sizable team, and you started a while ago.  That’s how you’re able to do all this stuff.  When did you start, and did it start as big as you are now?

[Connie]  [laughs]  No, definitely not.  So we started four years ago, and at that time, my cofounder --who’s Joe, he’s our CTO – and I both were working full-time, and we had a prototype for our first robot, which later became Hexy the Hexapod.   And we put it on Kickstarter, because we weren’t even sure if this idea of making robotics easy and affordable is actually popular, just at that time.  This is before the whole hype for robotics or STEM occurred, and so we weren’t really sure.  And so we put it on Kickstarted.  Our initial goal for that was fifty units, $13,000.  We ended up blowing past that on the first day, and by the end of the campaign had raised $170,000.  And so Joe quit his job.


[Connie]  I spent a little bit more time, because I also wanted to finish my projects that I was working on, because I was working with clients in STEM.  But then I quit my job, too, and then, you know, started doing this full-time, and, yeah.  So it – You know, we really heard from the customers that this was something that they wanted, and then we started working on our second project after Hexy the Hexapod was shipped.  Sparki really grew very much out of what we heard from our first Kickstarter community.  A lot of – Hexy was designed specifically for the hobbyist market.  It was designed for users fifteen and above, who already had some introduction to programming and are doing advanced programming and advanced robotics.  But a lot of them were parents who were like, I really – You know my kids watch me build Hexy, had a really great time.  Do you have something for them, because there really isn’t anything out there.  So we created Hexy [sic; Sparki], also did a Kickstarter.  We raised $190,000 there.

[Pius]  Wow.

[Connie]  In essence then, it’s been really explosive growth.  Now we are carried by over twenty-five distributors worldwide.  We’re in over 1000 schools and used by some of the best universities, like Stanford.  Also UT Austin used it at one point.  Purdue, Northwestern, MIT.  So it’s been really great to see.

[Pius]  So Kickstarter was pretty pivotal in starting Arcbotics, wasn’t it?

[Connie]  Absolutely.  For hardware companies, access to capital is the first big question.  Also you want to do something that is validated by your market, because there are a lot of inventory costs up front, a lot of R&D.  You have to purchase a lot of special equipment.  For us, we had to get a huge lasercutter in order to prototype Hexy, and that is not cheap.  And then 3D printers.  Now there’s a lot of options, but back then there were only a couple of really good options to do it, and that was very expensive.  Before you invest all of that research and money, you want to see if this is something that people want, so it was definitely very pivotal.

[Pius]  I can tell by what you’re saying, someone in your team has business experience, market validation and all that stuff.  Did you come from that world of raising capital and knowing how to run a business?

[Connie]  Yes, that would be me.


[Connie]  Yeah, so before this, I worked at a consulting firm called the Bridgespan Group, and they do strategic planning and business planning for nonprofit foundations and philanthropists.  And specifically they look at strategic initiatives, how to increase social impact.  And I focused specifically in K-12 STEM education, in addition to other things.  Bridgespan works with some of the best foundations and nonprofits out there, like Gates, and YMCA, so that was just such a fantastic learning experience for me, to learn how to do this.  But definitely I would say most of what I do now has been trial by fire…

[Pius]  [laughs]

[Connie]  …or just me Googling, what is this thing this retailer asked me to give them?  I’ve never heard about this before.  So there’s been mostly that.

[music interlude]

[Pius]  So what makes Arcbotics unique?

[Connie]  I think our focus has always been, one, open-source.  The new companies that have come don’t have that philosophy.  For us, that’s been really pivotal from the beginning.  In fact, we had to walk away from a lot of investors who were interested in our company because of that open-source aspect.  We didn’t want to budge on that, so that’s the big one I would say.  The second one is: making sure that we had excellent documentation and full tutorials.  There’s also a lot of products out there that might have some cool features, but then there isn’t any documentation, or especially lessons for teachers.  So it’s not ready to use in the classrooms.  And also we’ve seen a lot of products that didn’t start out for education, but have since pivoted to that.  They started as a really cool consumer gadget, and then transitioned to STEM when they maybe saturated that market and/or weren’t successful in that market, and then later just sort of put up a new landing page and said, now we’re STEM.  So there’s been a lot of that, as well.

[Pius]  So that brings up several questions, actually.  One, I’m wondering, is: who writes those educational lessons?  Do you have teachers on your team?

[Connie]  Yes, we’ve worked with different teachers over the years to create the whole set.  People who specifically focus on teaching STEM.  And we, ourselves, too, also wrote some of them.  More Joe doing that, because Joe, who’s our CTO, has a lot of experience teaching STEM and Arduino.  So prior to Arcbotics, he used to teach intro to Arduino classes at a Makerspace out in Boston, which is actually where we started originally.  And so, yeah, working with a whole bunch of people as well as ourselves.

[Pius]  Yeah, that’s great, because what I hear a lot from other businesspeople, like yourself, and teachers is that there’s oftentimes not communication between the two.  And so lessons that a business might write might not be in the best condition.

[Connie]  Yeah, exactly.  We’re really focused on creating good educational outcomes, social impact, not just, you know, more units. 

[Pius]  So why is open source, that open source movement, important to you?

[Connie]  We just believe that if you buy a product, you own it.  And you should own it for life.  And you should be able to do anything you want with it.  And seeing some of the projects that, let’s say parents or educators have done with our products, have been some of the most fun things to see.  Like for example, the University of Boulder, Colorado.  There’s a professor there, professor Karel, who had his students create a telepresence robot with Sparki.  And they took down the head and strapped on a smartphone, created their own software, and had the smartphone – they can visualize what the Sparki was seeing as it was going through.  And so it’s just really cool for us to see things like that.  We don’t want to limit what people can do.  And also it facilitates a lot more learning if our community can talk to each other, which is definitely what they do on our forums.

[Pius]  And originally – I know we’re jumping around here – but I wanted to ask you: Why was robotics education even important to you guys in the first place?  What prompted the Kickstarter?

[Connie]  I guess that question depends on how far back we want to go.

[Pius]  Sure.

[Connie]  When we were five, or when [laughs]

[Pius]  Well let’s start with the simple answer, maybe, right before you started the Kickstarter, and then I’ll ask you about the childhood questions.

[Connie]  [laughs]  No, I was born in…

[Pius]  [laughs]

[Connie]  So no.  Both of us came from different places with that.  So for me, my research and my work that I did at Bridgespan really showed me that, one, it’s a very important issue, and two, there are some fantastic providers out there, especially on the nonprofit side, who are trying to fulfill that gap in STEM education.  But when you look at the kinds of technical products that are out there, especially in hardware, they’re not designed for teachers with that in mind from the beginning.  And there were only a few big companies out there doing it at that time, like LEGO, Parallax, Pololu, but there weren’t – Arduino was already really popular at that time.  Robotics was starting to get popular.  There wasn’t any Arduino-based open source robots.  And so, yeah, so for me, it was seeing that gap.  And for Joe, he’s loved robots his whole life and wasn’t sure exactly how to make that more of a reality.  So he had been tinkering on the side from his job with Hexy the Hexapod and had developed it, but wasn’t sure how this would turn into a product or a company, especially an education-based one.  It was a really good partnership.

[Pius]  So it sounds like, for Joe, he did have that childhood drive to STEM education in the first place.

[Connie]  Yeah, so it’s interesting.  Both of us professionally have very different backgrounds, but where our interests stem from as a child was very similar.  So both of us were super nerds.  Super, super nerds growing up.  But both of us were fairly limited by how far self-teaching could go, because of our geography, or what was available to us at our schools.  Like for me, I loved math.  I loved genetics.  I loved programming as a kid.  But I grew up in Kansas.  We didn’t – I went to an excellent public school.  It was really great, but it just didn’t have a lot of resources, and I capped out very early with what I could do.  Like, for example, I tested – I started testing out of math classes when I was in fifth grade, but by tenth grade I ran out of classes to take.  And when I tried to learn programming, I became really interested, and that time PHP was more popular.  I was all, I want to learn PHP.  There weren’t any programming classes in the high school, so I had to go to the local community college, and there was one.  My mom scrounged up enough money for me to enroll in that class, and I loved it.  I remember thinking loops were really magical, but, you know, it was really difficult to pay for that one class, so then I just didn’t continue, and so it just sort of stopped there.  And for Joe it was even worse.  He grew up in very rural Maine, and the city where he grew up, now the population is 8,000 people.  He had even fewer options available to him than I did, and he had to try to teach himself everything.  And so for both of us, when we sat down and thought about what kind of experience we wanted to create for our users, it was really about – We want them to sit down and be able to, one, teach themselves or have a parent sit down and teach them with our lessons, without having had that experience.  Sorry, without having had that knowledge themselves.

[Pius]  Right.

[Connie]  And they can also progress.  It isn’t just something that they’ll pay for and then an hour later they’re done.  It’s something they can grow into.  We didn’t have that.

[Pius]  Yeah.  Are you planning more products, more robots, that will help kids learn programming and engineering?

[Connie]  Yeah, so, I mean, I think Sparki right now, we designed it to too well. [laughs]  We have just such a range of lessons there.  Our focus for Sparki is to expand the curriculum and expand the software available for it.  We’re not thinking about doing a different Sparki per se, because there’s really no need.  We just see this as an excellent product, and there’s so much to do there.  On the hobbyist side, we’re actually planning a launch for our third robot kit.  It’s going to be a humanoid robot.

[Pius]  Wow.

[Connie]  And the goal of that one is to bring the $15,000 humanoid robot capabilities down to about $500. 

[Pius]  Wow, that’s incredible.  And that’s coming up how soon again?

[Connie]  This fall.

[Pius]  Wow.  When I was at UT to get my Masters, I was trying to do humanoid robotics, and I can imagine the expense that goes into making something like that.  Do you have any strategies that you can share?  I know it’s proprietary, but how are you going to do that?  How are you going to make it cheap and accessible?

[Connie]  Well that’s something that is our specialty.  Basically taking very popular platforms that are out there, rebuilding it from the ground up, using a lot of design for manufacturing techniques, and also open sourcing it to do that.  It’s just basically what we’ve been doing from the beginning.  For us it’s not too difficult now only because we already – like for example, we already have all our 3D printers.  We already have a lot of the machinery that we need for that, because we invested, so it’s not too difficult.  It takes time to get it.  We were working on it now for almost two years.

[Pius]  I guess along those lines, what are some of the biggest challenges you’ve faced in bringing this company up or getting your products used?

[Connie]  I think two big things.  One, I mentioned earlier, is the access to capital part.  So after our Kickstarter had launched, we had a lot of investor interest, of course, from seeing that crazy growth.  And also we nailed a lot of retail partnerships early on, and a lot of investors were interested in that.  And we thought, OK, yeah, this is very expensive to do as a hardware company.  How do fund the next production batch? How do you do R&D for all these things?  But when we talked to investors, they didn’t like the fact that we were open source, and we were not going to budge on that.  They did not like that we gave away all our curriculum for free.  They didn’t like that either, because curriculum is something that is often charged for very expensively.  We didn’t believe in that, so we had to walk away from all that, so I think that was really hard for us in the beginning to make that choice.  We just had to be really creative.  We went to Kickstarter a second time.  We moved to Shenzen for a year and a half.  Also for manufacturing supply chain, but also definitely to save on the cost of living here. 

[Pius]  Sure.

[Connie]  Because we were in Boston, right.  So that was a fraction of the cost in Shenzen.  But you know, we had some fantastic retail partners who worked with us early on, and really believed in what we were doing, so had some great payment terms.  Just being really scrappy, and being very focused.

[Pius]  I think there are a lot of other people who are really into the open source movement, too, and if they are an entrepreneur similarly trying to get capital, do you have any tips for convincing people of the value of that?  Are there tips that you can share?

[Connie]  Well so we didn’t take investment.

[Pius]  Oh wow.

[Connie]  Yeah.  I don’t know if we are a good example of that because we never too investment, because we couldn’t find the alignment in the mission, but there’s such a wide variety of investors out there, that I would say, maybe look beyond – look for an angel who really aligns with your interests first.  Don’t worry too much about things like the exit.  Just really focus on creating the good product and having someone who believes in what you’re doing.  We’ve had people, friends of ours, who we’ve seen be successful with that route.  But for us it just wasn’t the right thing.  We wanted to do things in a very specific way, because that’s what we heard from our community.  That’s also what we personally believed in, and yeah, it’s worked out.

[Pius]  Yeah, that’s great.  So that second challenge that you faced, what was that?

[Connie]  So, I would say, letting more people know about us, because a lot of our competitors who are a lot more well-funded are taking out expensive ads across every single channel, and they’re letting people know just by sort of spamming all these different ads.  For us, it’s a lot harder, because we’re a small startup, and so figuring out, you know, how am I going to get the word out – because a lot of times teachers aren’t able to compare our product against the competitors because they don’t know.  So for us we rely heavily on word of mouth and strong referrals from our existing community.  Our existing community is also very active on social media.  They’re tweeting, they’re posting Facebook photos, they’re putting up YouTube videos of all their products, so that’s really great.  I would say that’s definitely a big challenge when you decide to go this route and stay true to what you want to do, but everyone else competing with giant marketing budgets.

[music interlude]

[Pius]  There’s an important issue that I want to raise with you, because you are in the engineering education business side of things.  There are a lot of teachers and educators who oftentimes a little distrustful of business or commercial entities, because maybe they feel like they’re always trying to sell them something, and they don’t really know how to educate.  Do you have any ideas about what might be causing this distrust, or ideas of how to mitigate this distrust between these two sides?

[Connie]  Yeah.  For me personally, I experienced the stark divide a lot when I transitioned into Arcbotics.  When I was at Bridgespan – Bridgespan themself, they’re a nonprofit that works with nonprofit providers.  So I spoke with a lot of teachers during that work.  And it was like, oh you’re one of us, because you’re on the nonprofit side.  But when I left to do a product to fulfill a gap that I saw from the research that I did, suddenly it was treated very – just did not want to speak to me, because they – oh you’re a business, oh you’re a company.  That has been really hard, and I think that perhaps – I think that this is actually related to the larger trend of how the kinds of businesses that people are starting have shifted, which is that previously, teachers would have dealt with very large, monolithic companies that have standard sales representatives that are spamming them with cold calls.  That might have been a very negative experience.  For us, even in 2012, but still even in 2016, the way that we’re running our company is new and becoming increasingly popular, which is, you can start it with a lot less capital.  You don’t need investment.  You can do a lot of these things.  So I think teachers in schools aren’t used to dealing with companies with our makeup.  And so still have a lot of that very understandable mistrust from a previous interaction.  I think the onus is really up to us to try to make sure that every teacher who interacts with us at every step of the way has a really positive experience and to see really what we’re about.  And so one of the new initiatives that we just launched this summer is offering a 60-day free trial to teachers actively teaching, who are interested in using Sparki in their classrooms.  It’s totally free.  They don’t have to pay anything.  They get a Sparki in person for 60 days.  They get to do all the lessons and all that, because we want to show people we’re really serious about our mission.  So that’s just one of the things.  But I don’t know.  It feels like a much larger of a thing that we’re battling.

[Pius]  Do you get feedback from those teachers?  That seems like an excellent way to build up some kind of relationship.

[Connie]  Oh yeah.  They’re very thankful.  It goes back to the earlier issue I mentioned about just getting our product in front of them at all, because we don’t have those really large marketing budgets to get in front of them, so they don’t even know about something like the free trial for Sparki at all.  I’m not really sure how to broach that.  And the other thing I’ve also seen, too, from teachers is, a lot of those companies, too, the large monolithic corporations that have been around forever selling these legacy education products, price the retail price so that it’s meant to be discounted significantly for teachers, because they never meant to sell at that price, because their margins are crazy anyway.  That was the complete opposite for us.  We don’t want to do a discount game.  We definitely don’t have nearly the same margins as they do.  Ours is a lot slimmer.  Because we want to make our products value-packed and affordable.  So it’s been hard for us sometimes when teachers are used to really large discounts, and I just have to say I’m really sorry.  This is the real price for this product.  We give you everything and all the lessons.  So I’m not really sure how to solve that per se.  We just are very consistent in our message when we interact with them, and we want to let our product and our community speak for themselves. 

[Pius]  You sound like the mom-and-pop robotics education store, if that’s a thing.

[Connie]  [laughs]  A bit.  I don’t think anyone’s ever called us that before, but…

[Pius]  In that description of the business, it’s like, you’ve got these big players, and then you’ve got, like, your new player, essentially, and you’ve got a different philosophy about things, so that’s what it sounds like.

[Connie]  Yeah, definitely.

[Pius]  OK, well I hope podcasts like this might promote the discussion a little bit more.

[Connie]  Hopefully, yeah.

[Pius]  So if teachers, who ultimately are going to be a big chunk of people who are interested in your product, if teachers want to get a hold of it, or get a hold of you, or the company, how can they do that?

[Connie]  Yeah, so they can visit our website at www.arcbotics.com, or they can contact me personally at connie@arcbotics.com.  Our website has a lot of information on Sparki.  We also make all the lessons available before anyone purchases so that they can see what they’re getting.

[Pius]  Cool.  That, I think, is it, Connie.


[Pius]  Thank you very much for talking.

[Connie]  Yeah, thank you so much.

[closing music]

[Pius]  As always, you can find the links to some of the items we mentioned today in the show notes or on the website: www.k12engineering.net.  If you liked this episode, please help me out, by rating or reviewing the show on iTunes, or sending me a message about your thoughts.  You can also post your thoughts on Reddit in the new engineering education subreddit.  Until next time, take care.


[Pius]  The views expressed in this podcast are our own, and they’re not necessarily the opinions of any schools, companies, or other groups with which we might be connected.  Our theme music comes from “School Zone” by The Honorable Sleaze.   Our closing music is from “Late for School” by Bleeptor.  Both are used under Creative Commons Attribution Licenses.

[music fadeout]