Transcript of Interview With Apurva Misra

Interview With Apurva Misra

This interview features Apurva Misra, chief executive officer at Competitive Consulting Advantage. You can watch it on Apurva’s profile page.

Transcript

– Hello, I’m Joe Welinske and I’m the conference director for Convey ux. And this will be our 12th annual event coming up at the last week of February. We’re both in person in Seattle as well as being online. And I have the great opportunity to speak with all of our presenters before the event. And so right now I am talking with Apurva Misra. Hello Apurva. How are you today?

– I’m good, how are you?

– Everything’s going well. It, it’s a, a really rainy day in the area north of Seattle where I’m at, which Seattle’s a headquarters blink. Where are you talking to us from?

– I’m talking to you from British Columbia. I’m based in Vancouver.

– Alright, great. Excellent city, wonderful place to visit. Well, I’m happy to be able to have you as part of this quick preview of conference sessions. Yours in fact, is going to be a workshop that is a half day in length. I’ll talk more about that in a, in a few minutes. But why don’t you start by telling us a little bit about your background and the type of work that you’re doing now.

– I’m a senior machine learning engineer at Truck Stop. I’m part of the machine learning team. We help truck drivers connect with shippers. Then you have brokers in the middle. It’s like the Uber of the trucking industry. Other than that, I do consulting as well. I consult for startups and I’m also teaching in a data engineering bootcamp in which we try to take non-technical people and teach them so that they can get into the data science field. Yeah, that’s about it.

– Well, and and how did you find your, your way into this? I I I think the area that you’ve, you’ve picked is, is one that maybe most people aren’t familiar with or understand how you find your way into that. So how, how did you get into this machine learning area?

– So after my undergrad, I started working in a startup, which was dealing with robotics. And that’s where I got introduced to computer Vision for the first time. And that was my like introduction to machine learning. And I decided to come for my master’s to Canada. And yeah, my research thesis was a little bit related to machine learning and then I like started like reading upon like books started doing like doing courses online and yeah, that’s how I got into the field.

– Well, the topic that you have is how to incorporate generative AI into your workflow, and that’s going to be one of our workshops, as I mentioned. So those are half day events that’s included in everyone’s ticket price and he has a lot of interaction, a lot more time to get into a particular topic. So why don’t you tell me a little bit about how you came upon developing this particular workshop and why you think it’s important for people to learn about?

– So when Chat g PT came out, I realized that that was the breakthrough moment when everyone got introduced to the power of AI and everyone like started using it, but it was hard for them to grasp on how to really use it well, what the limitations are and how, how, how well you can use it. So that’s when I started like looking into doing workshops and like going and presenting in conferences, especially for like people who are in tech but not so technical or people who are non-technical, helping them out, like grasp how how the AI works in there, but not go into the math as such. And it’s very important, especially when you’re working with AI to understand the biases and also the limitations. It would not work for all the cases. It’s based on how the kind of data you, you teach the AI with how diverse it is or how limited it is. It’s not the data of the whole world, right? Like you, you limit it by region or you limit by language. So it’s, it’s important to know the limitations and that is how you can use it well in your use case. And it’s, it’s really important, especially with generative AI because it generates stuff. It could be images, text, video, audio to make sure that you use your critical thinking before you publish it. Yeah, that’s, that’s why I think it’s important. And the workshop covers everything, like from the beginning, like trying to explain how the AI works, what’s behind it, what are the different use cases, different modalities like audio, video images and all the new updates. Like every single day this, this technology changes so much and also like a bunch of activities like interspersed in there. So it’s more interactive. Yeah.

– Well you’ve been doing, putting a lot of work into this workshop and you’ve done it with people before. Are there any particular challenges that you find come up, you know, people’s experiences in terms of what they find difficult to understand or the, you know, the parts that are maybe the most difficult to bring into their, into their work life?

– The part, especially like you mentioned, like how I have done it before and what are the problems that I’ve seen, like people face it during the workshop especially is to deal with like the mathematics in there. Like how does all of this data out there on the internet, the videos, images, the text, my comment on a random YouTube video, like how all of that is like, like converted into like numbers and thrown into a model to like train it, to like generate something as an output that I feel is like harder for people to grasp. So it needs like a lot of visualizations there, but I think that’s important, right? Like you need to understand it’s not even just about like the mathematics of the training, it’s also about the data. It’s, it’s so much more about the data that you’re putting in. So that’s very important.

– Well, it, AI is one of the major themes of, of our conference, so it’s great to have this workshop as part of that. Let me ask you, as you know, as someone that works with this, you know, yourself for your organization and other clients, you know, what’s, what’s one of the biggest challenges that you have as an experienced person? Are there certain places that you still find difficult or complex to be able to find solutions?

– I think especially with, with with the industry that I’m working with, it’s, it’s not that technically advanced, so we don’t have the data organized really well and with a lot of like, like my consulting work as well when I do it with the startups, there is a lot of data that they have, but it’s not organized well. It’s like, like mixed up and not structured. And so a lot of work goes into fixing the data before you can do anything with like machine learning that, that I feel is always a problem in, in the industry. Yeah.

– Well in addition to, you know, all of many senior leaders, experienced practitioners that we have at the conference every year, we also get a lot of new people coming in who are relatively new to the profession. You know, as someone that’s been doing this for a while, do you have a tip that you might share with new people of maybe things that you wish you’d known when you were first starting out?

– So to be honest, like I’m not from the UX field, but as a technical person, what I would say is if you’re getting into the UX field, just make sure you’re looking around and seeing how the technology is developing and keeping a tap on it. Like seeing how you can use it well to make your output better. That’s really important. It’s, it’s, it’s not okay to be afraid of it. It’s, it’s, it’s important to embrace it and figure out how you can use it. Yeah.

– Well it’s been great to have this preview time with you and thanks for sharing your thoughts and ideas with us. We’ll look forward to seeing you putting on this workshop in Seattle at the end of February.

– Thanks so much, Joe, for the invitation. I’m looking forward to it.

– Thank you.