Software systems to innovate and grow

Episode 36

Jon Marston with English Central: How Radio Hosting Led to Software Engineering

Show Details

Jon Marston, with English Central shares his experience of how contradictory backgrounds have led him along an untraditional career path.

As a college student, he began studying physics. During college he hosted a radio show because he loved music. And after college, he ventured in to radio news. Today Jon Marston is the VP of Engineering at EnglishCentral, an English teaching software program designed to teach English as a foreign language.

At first glance Jon’s career choices appear as opposites, but as you listen further, he will reveal how his creative background has in fact supported his journey.

Show Transcript

Carlos: Alright, Jon, my friend, thank you so much for coming on the show. How you’re today?

Jon: I’m doing pretty good.

Carlos: Alright, so Jon, in our pre-interview, in our previous conversations, I think one of the big things that made me think this would be an awesome subject is how your own background or your own experience somewhat converted to what you’re doing today, so let’s start there. Tell me a little bit about your background and tell me that story of how did you end up in software?

Jon: Well, I guess like all software engineers when I was a kid in junior high my parents had a computer and so I used to write little programs. I think I wrote a program when I was in 5th Grade to keep track of my comic book collection. Apple 2+ and always that was around when I was in high school, you know, did some computer classes and things like that. And when I was in college, I went to Cornell in upstate New York and I decided, this is the early 90’s or so, and I decided, wasn’t really interested in just working with computers. I want to explore the whole wide world of engineering and so in college I studied Physics, Engineering Physics which was pretty rigorous, difficult field with a lot of problem sets every week, at that time very little computing, very little software. I think I took one software course my whole undergraduate academic career was about numerical recipes in C and trying to solve Physics problems. But it was also the only course I got an A in. It was in my major everything else was not so great. Really I was just learning to be an engineer. I’m learning to solve problems and kind of learning methodology for how do you take a problem that is presented and do research, think about methodically. How you do solve the problem and how do you go from there? The other thing that I did in college was I spent a lot of time at my campus radio station.

Carlos: And that’s why you have such an excellent radio voice the way.

Jon: It’s been a while. But yeah, I used to do the 70s at 7, the Friday nights. You know the whole thing. What I found as that I really love music. I really love being a DJ and playing, you know, doing this 2AM shifts, doing these overnights and evening shifts and playing all kinds of music. I started exploring the news side of radio and loved it I guess is what I found. I was really passionate about it. I would spend nights hold up in a studio we had really real machines on those days and just place them together. Concert ads for Maceo Parker and Bootsy Collins and all kinds of crazy stuffs that I was really passionate about.

Carlos: There is contrast there between your two things. What do you think made you so interested in this super contrasting creative field versus what you’re doing before?

Jon: Well, it was something that I always wanted to do and I found out when I was growing up I enjoyed doing creative writing and as well as doing math. So it’s two sides that I just have always kept going through when I was a kid, even now I guess. And I know that when I was seeking out in engineering school I was very much drawn to a place where I would be exposed to more ideas than just engineering. I want friends who are artists. Who were going to be doctors or lawyers, or creative types and things. I wanted to really explore that and I guess I found over the years that that’s help me be an engineer. I got the rigorous training on how to solve problems, you know, very methodical way. But sometimes that’s not the right way to solve the problem or sometimes it’s not the right way to communicate a solution, and so having people that you are used to interacting with and working with that have another approach. You know, left brain, right brain different ways of thinking of things has definitely been a big plus especially doing things like user interface design or working with graphic artists or user experience people.

Carlos: Let me jump on a bit of a tangent and this is probably inner geek in me, so do you still like, I’m super interested in things of like Astrophysics and this sort of thing. It is from a scientific perspective but really a lot of it goes like above my brain and it just goes over my head very quickly. Are you still interested in this sort of subject as kind of this personal time?

Jon: I am. I love Nova. I love reading about the search for and dark matter and dark energy and things. But I will admit part of why I, you know, when I got out of school I didn’t pursue the PhD, it was also over my head.

Carlos: This is coming from your study.

Jon: I mean, I really love Quantum Physics and Quantum Mechanics and learning about Electrodynamic theory and all kinds of stuff like that. But I could see that this was the hardest of the hard. And people I was competing against were really passionate about just his. This was the only thing that they wanted to do and it was such a difficult field. I found that my passion was at that time in a much more creative field. And so it was clear to me I was never going to be able to compete with these guys.

Carlos: And just as an offside topic here, is this something that is made me think a little bit like the whole competition thing. I think that comes as kind of dumb in a way because yes some people are better than others but that gives us this mentality of scarcity like, “Oh, we don’t have enough.” We don’t need them anyways. Yes, maybe you’re not in the top three but even if you’re in the Top 100, you’re still in the 0.01% of the population that gets that sort of thing.

Jon: Well, yeah, I was comfortable with myself that was kind of where I was hitting a limit on what I could do. For me really the passion was part of it, and I’m confident that if I’m didn’t that passionate I would have chosen to pursue that. One thing I like about the industry is that there are so many different opportunities in so many different places in all kinds of variation of what you’re going to do. Academia, you know, so the place where I’ve been was sort of feeding into being a Physics professor and feeding into PhD program, to post grad programs to professorships and things. And that’s a really brutal competition where the number of people that get waddled down at each step along the way is very brutal. It’s a very very difficult track to do and it really just feeds in one direction so much of it. Whereas with private industry which is where I spent my whole career there’s so many vast different types of opportunities that people from all kinds of different backgrounds and all kinds of different things can do can jump in. I’ve certainly been happy and have never really regretted not pursuing the Physics track.

Carlos: Alright, so now let’s jump a little bit into like your professional life now, so you work in EnglishCentral. What is EnglishCentral for those that don’t know it?

Jon: EnglishCentral is a language learning site, if you know Rosetta Stone or things like that. And we’re focused on teaching people just English because that’s really the language of business around the world. Our core product is video and speech recognition system where we’re giving people videos to watch, things that they are interested in, things like movie trailers and different things like that. And then we give them the transcript to read along to and then we grade their pronunciation so we have a custom speech recognition type of system which listens to what they are saying and then gives them guidelines should they work on their L sound or their R sound and things like that. We also have vocabulary quizzing, and learning and connecting people with one on one human tutors as well.

Carlos: So just for us to get kind of a glimpse or sense of how large your teams are? You’re the VP of Engineering, how many people are in financial engines? How many people work their?

Jon: Overall we are about 400. Many of those are English teachers who were doing tutorials sessions. My engineering staff right now is probably around 25 and we’re globally distributed. Much development in the Philippines. I happened to be in Boston in United States for the team here. But we also have people in Japan, Korea, Turkey, Brazil, you know, our market is global and so we have a global presence and also engineers are distributed in different places.

Carlos: Alright, so now what is a VP of Engineering do at EnglishCentral? And I asked this question because I’ve seen this role all throughout the industry. But to me it could mean something different in different companies so what does a VP of Engineering do at EnglishCentral?

Jon: Well, my role is to build teams and to make sure that they can solve problems and that they are working on the right problems and kind of constantly be checking. Everything, be checking all the groups to make sure that they are all heading in the right direction and that they are not stuck. And then also be making sure that I know what they are going to be working on next. And so I was one of the founders of the company so we start off at just a handful of us and at the very beginning I think the most important thing was to understand architecture. So understand software architecture and to be able to jump in and write code and find really good people who could execute the vision, and since then the company has grown. You know, we split up into more and more sub teams. We have an iOS team, we have an Android team, we have a web frontend team, we’re backend REST services team, we have Quality Assurance team, we’ve all these different teams that are interacting as well as the designers and some of the different folks there. A lot of my goal, a lot of my job in the day to day is to be making sure that the right people are talking to each other. That somebody to know what everybody is working on and to know when, “Oh, you are working on this part of the system and you’re stuck on a problem. Why don’t you go over and talk to that person who has seen something similar before”, and you just facilitate communication; that ones are a huge part in my day to day job.

Carlos: So that means you have to have a ton of context in basically everywhere on a sprint to sprint level.

Jon: Yeah, that’s my job actually, is to understand how all the pieces fit together, where we want to be 6 months from now, where we want to be 3 months from now, and what are all the intermediary steps to get there, and to be feeding people the things that they can work on. Make sure that all the different sub teams understand what the overall goal is, understand what the overall vision is so they can make decisions along the way to get there. But it’s my job to really understand the whole global context now all the pieces fit together and why, and then to be watching things and just to be giving small corrections. The best is just to be giving small corrections. The lighter my touch can be the better I’m doing my job is one of the things I found.

Carlos: I read recently an article about this guy from I think Scribber, One of these companies, financial companies. Not Scribber, but I forgot the name of it now. But he was talking about how a COO does a lot of editing and not writing. And of course he means that as an analogy to this sort of kind of management overall that his role is particularly not telling people what to do but seeing base on what they do. He gets to edit and help them shape up base on their actions. It is somewhat of what you’re saying, I mean, it makes me think of that. But I think one thing that I just thought of that kind of puts two dots together is this whole notion of growing into a role, right. And just thinking back into about your story how you work from engineering to radio back then into software engineering. I think of how you grew into the software engineering role and how the kind of the little bit of the gap between the two, right.

Jon: Maybe I can tell you the, so I’ll explain how I went to school and loved radio. We’ll rewind a little bit, so I graduated, I’m going to date myself precisely right now. I graduated 1995 from Cornell with a degree in Engineering Physics. Then I asked my parents, “Hey, I just got this great Ivy League degree. I’m going into news radio.” They kind of put their head in their hands and then said, “Ok, let’s see how this goes.” And what I did for the next 10 years was work at a radio station in Boston called WBUR, just part of the National Public Radio Network. And I went in at the ground level as an audio engineer, you know somebody working the boards, placing microphones, doing mixes and stuffs like that, and just went pure radio. Understanding the technical thing was kind of a good piece they brought to it. And I worked my way up to be a producer, producing stories, which was a lot of fun. It was a chance to see the world. I had a microphone, had a recorder and went all over the world recording stories, meeting people, doing interviews, and then bringing them back, mixing them. And this was also around the time that the transition was happening in the industry and really in the whole world for all media, from analog to digital, and I get to be part of that and was part of launching some new shows. Some shows are still on. There is a show called “Here and Now” trying to help launch. There is another once called “On Point” which is also still on those both on National Public Radio. And we were trying to figure out how do we go from real tape to a whole workflow where we’re recording into computers and editing on desktop machines and getting the sound from place to place. And it was quite an experience, so it was quite a problem solving experience where the answer is going to be was not known. It was also very much like startups where you’re starting, you’re walking in, you’ve got a blank whiteboard and you’re saying, “Ok, what we are going to do.” And those experiences were great and I use those today. I use that same type of urgency. Nothing is more urgent than a live daily show in terms of deadlines and things like that. In software there is a world of sleeping deadlines that people have kind of grown accustomed to but when you’re on the air at noon and that is happening. You have to make sure that you make your deadline. That was a fantastic experience and I think like being a director in that media role is actually how I approach my VP Engineering role. The director is the one making sure that everybody, you know, the voice talent, audio engineer, the different people, they all have in front of them the thing that they are supposed to work on right now. But the director’s job is that be thinking about what’s next and what’s next after that. The director’s job is to also be thinking of how all those separate pieces are going to assemble together into one coherent hole, and that was really good experience. That’s how I approach my job now.

Carlos: A little bit to go actually, I was thinking of this and it just slipped my mind but basically it seems like you grew into that role, right? What is your opinion or what is your philosophy about how now some of your engineers are growing to roles, right? Or do you usually look for people who are able to grow into these roles in the same you did, again, by the mixed experiences and all these things that you had.

Jon: Always. I’m always looking for people that are looking to grow into the role.

Carlos: Why is that important?

Jon: Because they will bring enthusiasm. They will bring fresh ideas. You know, I would much rather hire somebody who is trying to… One of my key interview questions when I’m interviewing candidates for jobs is, “What is the technology that they are interested in working on that they haven’t had chance to work on yet.” I asked about different open source projects that they might be following or have contributed to. I want people who are passionate about what they do or generally interested in learning and growing. And that enthusiasm that people have for the field in general will translate itself directly into how they approach their job and what they invest into the product. Do they care about the quality of the code? Are they excited about working with some of the more senior people or great teachers so where they can learn their craft. People that I found and I have a harder time, one of the warning signs of somebody who is potentially a job candidate is if they’ve been working in the same exact technology for a very very long time. When I see somebody who has been strictly a Java Engineer for 10 years in a very precise part of it, part of the stack, is a warning sign to me. It’s a warning sign to me that they learn their one skill. They are not very ambitious and they are happy just kind of writing it out and until they are done with the technology. Until that technology is just out of date. People what I do like, what I’m looking for more old and more experienced people. I like people who are part of that turnover of the technology which is natural. Every 4 or 5 years there is a new way kind of doing things. These days it’s all JavaScript in the frontend. A few years ago it was Ruby on Rails.

Carlos: Flash before that.

Jon: Yeah, Flash before that. There is a whole reinvention that is critical to the technology so it’s not really about your age. It’s about do you like to continue to learn, and those people I find are really the best engineers. The ones who will invest the most, who will care the most, who will really, you know, when you’re on a hard project they are going to take that extra bit of pride in what they are doing. I also know that as a VP Engineer I am trying to get these teams going and the best is where I can get a team that has its own momentum. And I can make decisions along the way that could make decisions along the way and where I kind of just mapping out a direction. The hardest team is where they are not motivated and they are just kind of working their way, they are just kind of trudging from milestone to milestone and it never gets momentum of its own. That gets more difficult. And I find that the people who are really passionate about their craft want to learn new things are the ones that will come to the projects with the most momentum.

Carlos: What I find interesting about this that you just said is that it’s kind of universal. This applies to anything. This applies to any role and yes we’re talking about technology but this could apply to let’s say somebody in marketing, somebody in sales. Like, why are you interested in this?

Jon: I mean, marketing is completely different than it was 10 years ago? Now it’s all driven by Facebook and Google Ads, as opposed to the magazine ads. It’s all driven by what is trackable. User interface design is very trendy. If you think about how everything looks now, you know, with the flat look as opposed to 5 years ago when skeuomorphic was kind of the thing. But if you look at one of the histories of these icons that cover the past 20 years, you could just see these trends and these cycles and it’s constantly reinventing itself. And, you know, so the graphic design, the code, I mean right now we have a Java backend and some of my best engineers want to use Scala to bring that in. It’s helped quite a bit as people have been working with more distributed systems. They want to do more functional approach to programming. That has worked itself very well. On the frontend we used to be LAMPP stack (Linux, Apache, MySQL, PHP) in our case and we’ve been in our process moving this to be a full Angular stack talking to a REST service back end and kind of reinventing that as it goes through. I think it can be tricky, I mean you can get someone who just wants to reinvent for the sake of reinventing. You don’t want to waste all your time rewriting everything all the time but you want to be in that rhythm of the technology cycles. I find here about 5 years is kind of what these technology cycles tend to be, at least they have been. And you want people who are enthusiastic about that and ready to kind of keep reinventing as they go.

Carlos: How do you keep a tap on what did you just said on when it is too much? When are we innovating for innovation’s sake? How do you keep a tap on that as a VP of Engineering because again you got milestones to hit, you do have corporate goals etcetera that because those are product goals that yield revenue etcetera, when do you say like, “Hey, too many ideas. We got to hanker it down and focus on product.” How do you do that?

Jon: When your product people, the key is you have a product person who has some patience but not too much. In my company I think we that we have a refactoring budget. And I can get away with doing, ok we don’t have any new features in this release but we have to cleanup this code or introduce this new thing. I can get away with some of that but not too much. And if I’m spending all my time refactoring then I’ve gone too far. If the product is not moving forward there is no benefit to the end user and you’re just cycling because you want to play the latest toys then you’re getting nowhere. You have to kind of find a balance and I found the best ways to have product people who are impatient. You know, product people who understand technology and can understand, “Ok what is the sequence? Alright, alright, you need to replace this with that and then we can introduce this new great feature to users.” “Ok, then I’ll have patience for that.” But then if I at the end I’ll say, “No, I actually really want to use this other cool trendy thing.” And you want that product person who is going to say, “No. We have to shift to the product. We have to shift to our end users.” And the engineers and the people in my role who are really trying to think of that whole context needs to be really understand what the product vision is and understand that at the end of the day you’re in service of the product vision. You’re not just writing codes, write codes for yourself. This is not a hobby project and I guess that’s the best way to find the balance. It just understands that yes the code and the technology is in service with the product and if you’re not advancing the product it’s a hobby.

Carlos: If the product is not making money it’s a hobby, something similar. Like if you’re not looking at the business side of what you’re doing then it’s definitely a hobby. Alright, Jon, we are nearing the end here of our interview but I have a couple of final questions basically to try and get some advice from you. So we’ve got part of the last few questions, the first one is do you have any books or any resources that you kind of you would recommend everybody to go and check out whether it’s personal, engineering related or based on any of the topics we talked today. What’s something that you might have us check out?

Jon: Well, let’ see, so I’ll go for one new one, one old one. So the new thing, so I consume a lot of web stuff, I love Hacker News and I love Ars Technica and those are two tech feeds that I just consume constantly. The Hacker News is just the wide world of technology and just kind of keeping an eye on that kind of gives you sense of what are emerging trends. There is a lot of noise in there. There is a lot of garbage. There is a lot of stuff that’s never going anywhere or becomes a big deal. Ars Technica, I find is a great site for just well written journalism in our field and that they cover the big industry moves. The big technology shifts and things like that. Those are the two sites I’d like quite a bit. If we’re going to pick books, I guess what made really and this is going back, what made really big impression on me growing up was the book Soul of A New Machine by Tracy Kidder.

Carlos: Alright, well I haven’t heard of that one.

Jon: If you like history this is a good one. It follows the birth of a product called, I think it was Data General. This is back in the early 80’s. This is sort of a mainframe computer, and it is a narrative story kind of going through the whole cycle of a product and these engineers. What drives them and the different personality types? And, you know, we are in a completely different world now from the technology then but is a very well written narrative. So I guess those are, you know, Hacker News is just basically constant turn of articles but if you want to sit back and read a good book that will tell good stories that is engineering base, Soul of A New Machine — Tracy Kidder.

Carlos: Alright, we’ll put that on the show notes. This is our second to the last question, what will be one, if you could pick, I know it’s hard to pick one but what’s the most valuable piece of advice you’ve been given that you could share with other VPs of Engineering that might be listening to our show.

Jon: Well, I guess think about your architecture. There is that old advice about that, don’t over engineer or don’t optimize early, I think is kind one of those maxims. And I guess the best thing to develop is a sense of what choices matter when and knowing that when you make certain architecture decisions earlier you will be going to be living with them for a long time. And so you want to be thinking about when is the right time to make the different optimizations. You don’t want to optimize too early, you don’t want to optimize too late, you know, when your whole site is ground to a halt. You’ve stopped all future developments just so you can refactor. But there are certain architecture choices, and sometimes has to do with what type of primary keys you’re going to use. I guess there are certain architecture choices that you can make early, be thinking about them a lot. You think about them, what are the things when you later down the road have to switch to one of thing to many of things, is better to make that choice earlier so you’re starting off with many things so that you don’t have to grind everything to a halt later. So if you can start on a database that distributes well as opposed to a monolithic relational database and you know that you can build from the beginnings that you were working from those constraints without slowing down your project to a halt and taking forever at your first iteration at the door. That’s what you want to be thinking about, so what to optimize when I think is the best skill to try and develop.

Carlos: Alright, I think that’s one I hadn’t heard of as of tonight there, and it made a ton of sense. You don’t necessarily need every choice at every given time. There is kind a time for everything. Cool. Alright, Jon, and now the last one and probably the most important, are you guys hiring is the question. And if you are, how can people find more about EnglishCentral and how can they get in touch with you?

Jon: Well, at the moment we’re hiring in Manila. So if you are listening to this and you are in the Philippines and you are in the Manila Metropolitan Area, look us up. Look up englishcentral.com, send an email and we’ll be happy to talk to you if you’re a technical folk. At the moment, so our team in the U.S. is kind of tied, you know, we’re not expanding our U.S. architecture in senior management. But we are looking for engineers that are passionate about Angular, passionate about JavaScript, Android development, iOS development and test engineers. The often overlooked extremely valuable person is the test engineer who loves scripting and is really creative at how to approach these problems for writing great automated tests. It’s a tough skill, always in demand and if you’re looking to break into engineering fields, you want to be a programmer but you’re having a tough time getting past some of these classic programmer exams that advance script driven test engineer route is a good one to explore. Opportunities will open up and you will find that there is not as much competition for those jobs.

Carlos: Alright, Jon, well my friend, thank you so much for your time today. And for everybody who are listening make sure you check out englishcentral.com and we’ll also put any contact information we may have for Jon on the show notes. So Jon, thank you so so much for being part of the show today and looking forward to catching you up soon.

Jon: Thank you!