At Ponder, we celebrate the diverse background and perspectives that each of our team members brings, so we started a “Why I Joined Ponder” series to share and learn from their experiences. Our second interview was with one of our Founding Engineers, Mahesh Vashishtha. Peter Olson (Chief of Staff) was the interviewer. What follows are highlights. You can listen to the full interview here, or read the full transcript here.
During the interview, Mahesh makes several references to the Ponder product — pandas on a data warehouse — which you can try here.
Why Mahesh joined Ponder, in a nutshell:
Mahesh grew up in Irvine, California, in a family more focused on biology than computer science. At some point he programmed his TI-84 calculator to play a few games, took AP Computer Science, and read Gödel, Escher, Bach, but it wasn’t until partway through his time at UC Berkeley as an undergrad that he committed to majoring in Computer Science. He then worked at Google as a software engineer for about four years before coming to Ponder.
So why did Mahesh join Ponder? In his words (lightly edited):
I was happy at Google working on search near the end of 2021. I had known Doris [a Ponder co-founder] in college. She was one of the first people I met at Berkeley and she was part of a group of friends that I’m still associated with today. So Doris reached out to me to chat. She told me about Ponder – that they had just started the company and what they were doing. I ended up talking to Devin and Aditya [the other two Ponder co-founders]. I was pretty excited. The main thing that was attracting me was the pace would be a lot faster. Obviously when there’s no product you iterate fast and try to make one.
I had thought of joining a startup at some point. I figured it would be a good experience. I wasn’t planning for it to happen that year but I thought the technology in Modin and Lux was really cool. It had good traction and it was coming out of interesting research at Berkeley. And then specifically I knew Doris. I figured I would like working with her. I really liked talking with Devin about technical things and with Aditya as well, and also about the strategy of the company so I figured these would be good people to work with and they would run the company in a good way. So I thought, if I’m going to join a startup this seems like a good opportunity.
With Modin I think what was interesting is you keep the same pandas API, but there are a lot of technical details in the implementation. It’s pretty subtle to parallelize it. Devin told me early on – handling metadata for example, that’s something that pandas has a lot of very particular behavior about the index and the columns of the data frame. You can have a MultiIndex on either one, and then a lot of operations will change the index. Supporting that in the distributed implementation is tricky. But what’s interesting is that and all the other problems that come with a distributed implementation are totally hidden from the user. So I thought that was interesting. And then with Lux too, I think I also liked the user focus. It seemed like something that I could see a data scientist benefiting from. And I also liked the emphasis on doing heavy work – technical, heavy work, but not exposing it to the user. So the user just gets something that’s useful to them.
My hope to have a faster pace has definitely come true. With most of Modin, you can get PRs merged within a few days, and then they deploy, hopefully within weeks, which is kind of similar to Google, but I think because Modin is in an earlier stage, you can make bigger changes in a shorter time. And then internally with the product at Ponder, I think we’re moving quite fast. So it’s good. I think at Google it sometimes felt like I was putting out a lot of things and it felt like I wasn’t really going that far or pushing the team that far. But here, it feels like I can actually make the team go far with the more effort I put in.
I would say one of the things I like here is that everyone’s doing something important. There’s a lot of work to do. And everyone is doing something that’s an important part of moving forward. Every single person is really interested in making the company succeed, and in helping each other move forward.
I’m excited about the product that is now available. Generally I think there’s a collision between data science and big data. I’m really hoping that Ponder will be an important part of that.
I keep talking about speed, but a lot of the things that I think are positive about Ponder relate to that. I think it’s amazing how quickly we can move from not having anything to having the solution to a particular problem. I think we have a very can-do attitude to solving engineering problems. And I really like that. And it’s great to work with really enthusiastic people. Everyone is very motivated to make the final product good.
———————————————
For those who want to listen to or read the full interview, see below for our Spotify podcast, as well as a lightly edited transcript:
Structure of the Full Interview:
- Part 1: Growing up in Irvine, CA; reading The Lord of the Rings; programming games on a TI-84; discovering Gödel, Escher, Bach
- Part 2: Attending UC Berkeley; picking a major; doing research and internships
- Part 3: Working at Google
- Part 4: Why I Joined Ponder, and advice for computer scientists earlier in their careers
Part 1: Growing up in Irvine, CA; reading The Lord of the Rings; programming games on a TI-84; discovering Gödel, Escher, Bach
Peter: Mahesh, thank you so much for talking to me today about why you joined Ponder. We’re going to ask a bunch of questions, learn a bunch about your life and finish with this question of: “Why did you join the company?” But first I want to start by asking about when you were younger, your early interest in computer science. Where did you grow up? What were you like as a child? Did you show an interest in computer science from a young age?
Mahesh: So I grew up in Irvine, California. As a child, I would say I was quiet and kind of shy. I liked reading. I used to read a lot of fantasy. I was a big Lord of the Rings fan. I even read some of the extended universe like The Silmarillion. I read Harry Potter and lots of Redwall. And I’d say I liked school for the most part, but I didn’t really like doing homework or studying. I think that’s pretty common.
Peter: So in an immediate aside from the task at hand, I want to ask you – With Redwall, I often say the word “vittles.” That’s a Redwall thing, right?
Mahesh: Yeah, it definitely happens a lot in Redwall.
Peter: I think the true word is v-i-c-t-u-a-l-s, but I don’t know how to pronounce it. In Redwall they always write v-i-t-t-l-e-s.
Mahesh: Yes. Yeah, they do.
Peter: Well, for all those listening about why Mahesh joined Ponder, you can go read Redwall and you’ll learn more. So you said you grew up in Irvine. When were you first introduced to computer science as an idea?
Mahesh: Yeah, I don’t know if I know about that. I mean, programming – I sort of knew you could program for a while. I don’t know when I learned about computer science and understood that it’s something that goes beyond just programming. I started programming when I was 13 or 14. My friend showed me how to program on our graphing calculators. I had a TI-84 and he showed me how to make a game that was just – It wasn’t really a game, it was just – There was a theta, a character displayed on the screen, and you just moved it around with the arrow keys. And so it must have been 10 lines or something. But from that I kind of figured out how to make games. I made a small Frogger game. Maybe something else. And I don’t think I even knew what language it was. There was a list of commands available. I guess it was like TI-BASIC. And yeah, so I kind of guessed my way through and that was how I got started. And then I had a little bit of more experience in high school. I took AP computer science when I was a senior.
Peter: Did you know any computer scientists? Any programmers?
Mahesh: Yeah, I guess I knew a lot. I must have had a lot of family friends who were programmers. But actually in my home, my parents are biologists. My dad’s a physician scientist, and my mom is a biologist by training. So we were always talking about biology and medicine. They didn’t know how to program of course.
Peter: So when you were in high school, did you have a sense that this was a career, something you could do every day all day long to make a living. When did that emerge? When did you start coming to think: “Maybe I’ll be a computer scientist?”
Mahesh: I think I might have thought about it occasionally in high school. But yeah, officially I started college as an intended biology major. So that was, I think, the plan when I was applying to college, but I also had an interest in computer science. Also, a little before I left for college, I read Gödel, Escher, Bach, which is a book about computer science, mathematics. I think it was a really important book for me, because that was my first exposure to more abstract mathematics and it talks about some fundamental ideas in computer science, and I got really excited about that stuff. So I had a sense that I wanted to do more of that when I joined college.
Peter: And you read that in high school – sort of near the end of high school?
Mahesh: It was right after high school. Summer after high school.
Part 2: Attending UC Berkeley; picking a major; doing research and internships
Peter: Let’s talk for a second about transitioning to college. So you went to UC Berkeley and you said you initially wanted to be a biology major. When did you switch from biology? How did you decide to switch?
Mahesh: Yeah, so I said “intended biology major,” because in the College of Letters and Science at Berkeley, you don’t start with a major. Your intended major kind of puts you with certain advisors who will help you get into that major and have a career there. I started taking computer science classes from the beginning, and I loved them. The programming was fun. But the exposure to computer science concepts like recursion. And for my first course, we built an interpreter in Python. That was a lot of fun. I kept taking computer science courses. I kind of just didn’t take enough bio. I took some chemistry, organic chemistry, but I was more interested in math and CS.
Peter: Did you ever contemplate studying something other than CS? Like, did you think maybe I’ll do math, instead of that? How did you settle on CS?
Mahesh: I was very unfocused, or I’d say mercurial, in college with my study and career plans. So I declared the CS major my second year, but I kept talking about switching to math or physics. I even thought about applying to med school. I took up other courses on the side. The CS major at Berkeley – I was able to get away with taking, I think, six computer science classes, and this is on a semester system. So that’s very few. I think very few majors would allow that. But you could use other kinds of courses, like electrical engineering, and even math for some of the requirements. So I took some EE classes, which were more like applied math. They were optimization, probability, random processes, signal processing. I did a bit of all that stuff. Very math-focused. Applied math, you could say.
Peter: How would you describe the CS / applied math culture at Berkeley? Did taking classes with your peers feel collaborative? What was that experience like?
Mahesh: It was exciting. I think a lot of the professors are really good. Many of them, the ones who were teaching classes, were at the cutting edge of research in their fields. I was having fun. I wouldn’t say my experience was very collaborative – I would mostly just go to lecture and do homework myself and study by myself. And also, because I was jumping around between all these different kinds of courses – I had a group of friends socially, but I didn’t really have a group of peers that I was taking a lot of different classes with. So you know, I’d meet some people in different classes and make friends there.
Peter: How did you decide how to spend your summers? I know that eventually you did research in undergrad. I know that you ended up interning later on for Google. What did you do your first couple summers as an undergrad?
Mahesh: The first summer I went back home to Irvine and I was supposed to be working with a biology professor at UC Irvine. They were doing wet lab research, I think with a worm – maybe C. elegans. They would take microscopic images of a worm – multiple images of the same worm, I guess, and then they’d have to stitch them together. So there was a program that used to do that as an open-source library. I was supposed to improve the performance of that library, but I didn’t really get that much done that summer. So that’s what happened in my first summer. As to how I decided that? I mean, I had a sense I wanted to do maybe biology and I liked CS. I thought that was something that was kind of a bit of both. And then the next summer I took a physics class.
Peter: Did you stay on campus?
Mahesh: Yeah, I was on campus.
Peter: At some point, you start doing research in controls. You talked about your interest in applied math. And you were working with Professor Tomlin. How did you get started with that? What was that experience like?
Mahesh: So Professor Tomlin taught one of my courses and I cold emailed a bunch of professors, and I didn’t really know her from the class. But I emailed her and I thought her work was interesting. And so I started reading some papers about what her group was doing. I worked with a really good mentor called Mo Chen. So he helped me get up to speed and then start doing the research. If you want, I can go into a little more detail about what I was doing.
Peter: Yeah. I’m interested in part because you were doing this and your current work isn’t directly related – I mean, there are foundational things that you learn from that, and you learned more about research – but I’m interested in part to know enough about what that experience was like to know: Why did you end up not pursuing that path more, going to grad school in the controls area instead? I think a lot of this is an attempt to contextualize this final decision of how did you wind up here doing what you’re doing today. I’m interested a little bit more about the controls work and what that experience was like.
Mahesh: For context, in optimal control, you have a dynamical system, which is some system that’s evolving over time, and you have some ways of controlling it. So this group’s research was usually applied – part of it was applied to unmanned air vehicles, like drones. This drone state would be its position in space, its velocity, things like that. And then you can control it maybe by making it accelerate in a certain way. They use a theory called Hamilton-Jacobi Reachability, which lets you define a certain set of states that you want to avoid, or maybe a set of states that you want to get to. And then you can try to understand, given those goals, what are the set of states that will take me there? What are the controls that need to reach there? So that’s said in a kind of abstract way, but it can be applied to do things like, say: You don’t want the drone to tip over, or you want to avoid an obstacle so how do you avoid it. How do I make sure that I can start my trajectory so that I can get to a certain place? There are partial differential equations you can use to solve and to answer these questions, but they suffer from the curse of dimensionality so they get really hard to compute as your state gets more complicated. So I was looking into ways to deal with that. If you have an equation in multiple variables to maybe split it into two equations, and fewer variables. Some of that was kind of like playing with equations. I did some playing with the actual computation in MATLAB. I had a lot of fun. I got to do some creative work and collaborate with good people. It was a good experience. I was working there on and off for maybe a year.
Peter: Can I ask a question? In this context, does “unmanned” mean that – it means obviously, that there’s no person inside the vehicle, but does it also mean that it’s autonomous? Or can it be remote controlled? In this context, you mean: No one’s in there and also it’s kind of choosing it’s – Maybe you give it an objective, but it’s controlling everything else itself in an automated way?
Mahesh: I would guess that’s the goal. I don’t know in practice, like how much of drones are automated.
Peter: I have a good friend from grad school who was doing a PhD in aerospace engineering, but was working on a very similar thing with getting a drone to land on an aircraft carrier. If you do a full simulation, you can do it, but to do it in real time, just getting something that can process everything fast enough to actually make it possible. This sounds super hard. So you did this. Did you ever seriously consider: “Hey, maybe I want to go to grad school in this area?” and then how did that conversation unfold?
Mahesh: I thought about it. I think I probably could have gotten into grad school. But I would say I was still a little too unfocused. I still had an interest in math, physics, software engineering. I wasn’t very particularly motivated to go to grad school. Yeah, and then I ended up doing an internship the next summer at Google.
Peter: How did that happen? How did you decide that’s what you wanted to do? It doesn’t strike me as the kind of thing that just falls in one’s lap. You’ve got to work pretty hard to get those internships. How did that happen?
Mahesh: I knew I wanted to get some industry experience. I think I applied to several places. I had a referral to get into Google. Someone referred me and that’s how I was able to get the interviews. So I prepared for the coding interviews with Cracking the Coding Interview. And I managed to get an internship.
Peter: So you do your internship, and eventually you go back there full time. Was life as an intern quite different from your experience as a full-time worker, or did the full-time work felt like a continuation of the internship?
Mahesh: I think the main difference is that you spend so much time getting ramped up that you don’t have a lot of time as a productive member of the team. I think when I joined full time at Google, I was working on a mature system of a product. I felt like I wasn’t really ramped up until a year later. That felt like the biggest difference. But otherwise, it did feel like I was kind of like a full-time employee.
Peter: What were you working on during your internship at Google?
Mahesh: Let’s see. It’s been a little while. There’s the Google Cloud SDK, which is a set of libraries that people who want to use Google Cloud in some context use to run functions on Google Cloud. So it spanned a lot of different projects, but I guess it included BigQuery and some other less high profile APIs. The client side API would be specified in a protocol buffer, which is language independent. It’s just a data storage format, which is used everywhere in Google. And then from that protocol buffer, you had to create client-side code in I think seven different languages. So there was .Net, JavaScript, Python and some others. The team I was on was working on actually converting this description of the API into code. It was generating code. I was working on taking the generated code and publishing it to GitHub. I think that was most of my work during the internship. Some of it was open source as part of this open source tool that was used to generate the code, I think. And then part of it was proprietary and lived in Google, because I think it was touching something internally and it had to be done within the Google source code.
Peter: In your internship, were you in Mountain View? Were you at Google headquarters?
Mahesh: Yeah, I was in Google Mountain View.
Peter: I’m interested for just a moment in talking about your location over time, and your sense of place. You grew up in the Irvine area, and then went to college in Berkeley, and then you did this internship. When you finished undergrad, did you go back to Mountain View? How did your internal sense of place change over time?
Mahesh: I went to Berkeley. I really liked being in Berkeley. It was very different from Irvine. Irvine is a nice kind of suburban city. I like Irvine. But Berkeley I like in a very different way. It was the first time I could walk around to stores and restaurants outside my house. And it was fun to just walk around the campus. Beautiful campus. And I definitely had a home there. I stayed in the same place for my last three years. I had a nice group of friends. So I felt grounded in Berkeley. There was a bit of a gap after Berkeley. And then I started work in Mountain View. Mountain View I didn’t like as much – It’s a little more spread out. But the Google campus was nice. It didn’t feel very closely connected to Berkeley. They’re geographically not too far apart, but a lot of my friends were gone. So it felt like a different stage.
Part 3: Working at Google
Peter: So then your life in Mountain View – You start full time. Did you switch teams? Did you stay on the same team? Talk a little bit about your Google experience.
Mahesh: I got hired without being on a particular team. I think that’s still how Google does new grad hires.
Peter: This is 2017. Is that right?
Mahesh: Yeah. I signed the offer in 2016. And then I joined at the end of 2017.
Peter: Okay, great.
Mahesh: I talked to a few teams and ended up working on ads. I was working on a team that was helping advertisers measure the impact of their spend. They were particularly interested in conversions, which is when someone who’s clicked on an ad ends up doing something the advertisers are interested in, which is very often buying something, but it could be signing up for a mailing list or something else. There’s a lot of sophisticated infrastructure on Google’s side to link the search to the click to the conversion, or multiple conversion events. So that was the focus of my team’s work.
Peter: I’ll put an idea out there and get your reaction to it. I think it would feel intimidating to me to step into that context because so much effort, energy, resources, time and attention from lots of bright people have gone into thinking about those areas. Did you get there and feel like: “Oh, there’s low hanging fruit, there’s well-scoped projects that will clearly push things forward?” Or did it feel like: “Oh, wow, how do I push forward the cutting edge of what’s already so cutting edge?”
Mahesh: Yeah, I’d say it was more like the latter. I was building on top of a lot of useful work, not just in ads, but in other parts of Google’s infrastructure. So all the way from the physical data center to the Borg, which is the infrastructure that a lot of things besides cloud run on. And then even all the internal tools they have for reviewing code and running tests – It’s all very sophisticated. So it definitely felt like I was making small incremental improvements. And then within ads, of course, there’s lots of mature technology that has been developed over the years and some of it’s quite good.
Peter: How did you feel about the day-to-day working arrangements, connection with your team – Culturally, what was it like to work at Google?
Mahesh: It was good. When I joined, I don’t think I had a lot of practical experience with software engineering or programming in general. I hadn’t taken a lot of practical courses for programming or software engineering. I got a lot of mentorship from people who were more experienced. One of the things I like about Google is that there’s a huge emphasis on engineering excellence. So things like writing really clean code. It has to follow the Google style guide. It can be painful when you’re getting started. But the result is that most of the code that’s checked in – At a high level, maybe there’s some issues. It needs to be refactored or something like that. But at the lower level, the code looks nice, and you can generally tell what a function does or what a test does. I think these kinds of standards about writing code – how to test your code, and then things like how to deploy code safely to production – I think these are really good things that I learned there. Day to day, I was writing a lot of code myself. I would write design docs. We’d have stand ups a few times a week. I think this may have varied over time, depending on which team I was on.
Peter: Just for a moment more on Google, before we move on to answering the question of why you joined Ponder. How much heterogeneity was there between teams? I know you ended up working on a few different teams during your almost four years at Google. Is there a pretty consistent feel and experience no matter what team you’re on? Or would you say actually, no, quite a bit of the day-to-day feeling of life did vary by team?
Mahesh: Yeah, for me it varied. But first, I worked on ads. And then I went to a smaller project called Data Commons, which is kind of new. Just a few years old. It’s an open data project. So it’s about gathering together open statistical data sets from many sources. Giving people a way to have easy access to them and to compare data from multiple sources. It was very different. It was a little bit more like a startup. Things moved faster. There was a faster feedback loop for deploying things. There was less burden to not mess up things that were already doing well. Then search and ads. Yeah, later I worked on search for a year. Things moved a little slower. You have to make sure when you push a change that it’s not going to mess something up. So there’s a lot of steps you take to make sure of that. I think that’s probably the biggest difference.
Part 4: Why I Joined Ponder, and advice for computer scientists earlier in their careers
Peter: You spent this time at Google and you had a bunch of experiences there. It sounds like the Data Commons one was a little more perhaps like what you’re experiencing at Ponder today where you have more flexibility, more autonomy. And then you went back to a bigger team, a more established team, working on search. At some point the idea of working at Ponder comes across the dashboard. Why did you join Ponder? Can you tell me more about that story?
Mahesh: I was happy at Google working on search near the end of 2021, or maybe middle of 2021. I had known Doris [a Ponder co-founder] in college. She was one of the first people I met at Berkeley and she was part of a group of friends that I’m still associated with today. So Doris reached out to me to chat. She told me about Ponder – that they had just started the company and what they were doing. I ended up talking to Devin and Aditya [the other two Ponder co-founders]. I was pretty excited. The main thing that was attracting me was the pace would be a lot faster. Obviously when there’s no product you iterate fast and try to make one. That would be nice. Specifically about Ponder – I had thought of joining a startup at some point. I figured it would be a good experience. I wasn’t planning for it to happen that year but I thought the technology in Modin and Lux was really cool. It had good traction and it was coming out of interesting research at Berkeley. And then specifically I knew Doris. I figured I would like working with her. I really liked talking with Devin about technical things and with Aditya as well, and also about the strategy of the company so I figured these would be good people to work with and they would run the company in a good way. So I thought, if I’m going to join a startup this seems like a good opportunity.
Peter: You said that you thought Modin and Lux seemed like cool tools. The technology was interesting to you. Can you say a little more about that. What about the project caught your attention?
Mahesh: With Modin I think what was interesting is you keep the same pandas API, but there are a lot of technical details in the implementation. It’s pretty subtle to parallelize it. Devin told me early on – handling metadata for example, that’s something that pandas has a lot of very particular behavior about the index and the columns of the data frame. You can have a MultiIndex on either one, and then a lot of operations will change the index. Supporting that in the distributed implementation is tricky. But what’s interesting is that and all the other problems that come with a distributed implementation are totally hidden from the user. So I thought that was interesting. It sounded like a good project in that way. And then with Lux too, I think I also liked the user focus. It seemed like something that I could see a data scientist benefiting from. And I also liked the emphasis on doing heavy work – technical, heavy work, but not exposing it to the user. So the user just gets something that’s useful to them.
Peter: And I should say, for all those reading or listening, you came to Ponder as one of the founding engineers. So there were the three founders and then I don’t know how quickly all four founding engineers came, but there were four founding engineers – Is that right?
Mahesh: Yeah.
Peter: So you’ve now been at Ponder for over a year. How have things changed in that time? Have your initial reasons for coming proven true? You were interested in working at a startup because of the speed. What has the arc of that year been like?
Mahesh: It’s been really busy, and I think it’s been good. So a lot of things have changed over the year or so that I’ve been here. When I started, we were mostly focused on developing Modin. So the other founding engineers and I were getting up to speed with Modin. Then we were thinking about strategizing about what kind of products would be useful. That was happening early on. And then we were more focused on design partnerships for companies that were using Modin. That was one big focus. And then earlier this year we started to get more focused on actually developing some of the initial products. I helped Devin make an initial prototype of the product you can find now on ponder.io/product, which is supposed to run pandas on your data warehouse. I kept contributing to Modin, and then making some contributions to the product. Overall, there’s been an arc from supporting Modin open source to working more on the new product and actually fleshing out the details of it. It’s been fun to see that and to be part of it.
Peter: Have you had to change your working style at all from your time at Google? What has stayed the same and what has changed for you?
Mahesh: I would say I work mostly in a similar way. My hope to have a faster pace has definitely come true. With most of Modin, you can get PRs merged within a few days, and then they deploy, hopefully within weeks, which is kind of similar to Google, but I think because Modin is in an earlier stage, you can make bigger changes in a shorter time. And then internally with the product at Ponder, I think we’re moving quite fast. So it’s good. I think at Google it sometimes felt like I was putting out a lot of things and it felt like I wasn’t really going that far or pushing the team that far. But here, it feels like I can actually make the team go far with the more effort I put in. In terms of the practical effects, I’m on Slack a lot. I am working with people in real time. Yeah, I like that.
Peter: What would you say to someone who’s considering making the jump from a larger company to Ponder specifically? What kind of advice would you give them?
Mahesh: Advice about how to succeed at Ponder?
Peter: Yes, both how to succeed, and also things they should know as they contemplate making the jump.
Mahesh: So right now, I would say one of the things I like here is that everyone’s doing something important. There’s a lot of work to do. And everyone is doing something that’s an important part of moving forward. One of the things I like here is that every single person is really interested in making the company succeed, and in helping each other move forward. That’s a good environment because I think personally having the feedback loop from when I start working on something to when it gets used, or we can get feedback about it, is really important. I think at Ponder, we try to make that happen really quickly. And I think that’s really good.
Peter: Okay, two more questions. One of them is: What advice do you have for current college students studying computer science about getting started in their careers? What are things you wish you had known earlier – hard-won understandings about what it means to be a professional software developer?
Mahesh: One thing I’ve come to understand is that software engineering is very different from computer science. I think some people say software engineering should be a different major than computer science. I don’t know where I stand on that. But I think it’s important to be aware when you’re entering CS or engineering that software engineering and CS are complementary fields. I spent some time studying CS, but I feel like most of the things I learned aren’t really relevant to what I do today, which I think is true in most fields. If you want to get practical experience, most university classes are not really good for that.
Peter: My last question is: What are you most excited about working on in the next year, within the scope of things you’re able to talk about?
Mahesh: I’m excited about the product that is now available. We’re going to add more features and expand the context in which the product can be used. So I’m really looking forward to seeing that. Generally I think there’s a collision between data science and big data. I’m really hoping that Ponder will be an important part of that.
Peter: Is there anything else you want to add – Important things about your “Why I Joined Ponder” story before we close?
Mahesh: I keep talking about speed, but a lot of the things that I think are positive about Ponder relate to that. I think it’s amazing how quickly we can move from not having anything to having the solution to a particular problem. I think we have a very can-do attitude to solving engineering problems. And I really like that. And it’s great to work with really enthusiastic people. Everyone is very motivated to make the final product good.
Peter: How does that enthusiasm manifest itself?
Mahesh: In other people? I think people will comment on everything you do – not everything, but if someone has any objections to something you’ve done, they’ll voice it. So we’ll get into discussions about that, which I think are good. And then sometimes people will come up with – for example, I probably post a lot of messages about improving our testing procedure for the product. And if something is bothering me about our development workflow, I will voice it, and then people will usually try to help me out. A lot of talking, I guess. We try to talk on Slack and asynchronously on Notion.
Peter. Mahesh, thank you so much. This is awesome. I’m so glad you’re at Ponder. Thanks for taking the time.
Mahesh: Yeah, of course.
