The conversation this week is with Fraser Gray-Smith. Fraser is a business strategy professional that specializes in using analytics to drive value in organizations. He takes advantage of his unique mix of technical skills and business acumen to uncover valuable insights, develop business cases, create management buy-in and manage the execution of projects. Outside of work, Fraser likes to learn about leadership and politics. He also enjoys experimenting with emerging technologies in the finance and machine learning fields.
If you are interested in learning about how AI is being applied across multiple industries, be sure to join us at a future AppliedAI Monthly meetup and help support us so we can make future Emerging Technologies North non-profit events!
Resources and Topics Mentioned in this Episode
Fraser Gray-Smith 0:00
I think that there's a proliferation of tools. There's a lot of different tools out there that companies use, that are generating data. And they want to connect all of these data points together, they want to bring data from lots of different sources into one place into one model. And I think that's the right thing to do. I think generally speaking, the more data that you can put into, especially some something like a neural net model, but because you've got data residing in all of these silos, one of the things that has become an integral part of the AI process is this data engineer. How do we take data from these 10 different systems and put it into one place and connect it all together in a way that makes sense? I think that that's the the addition to cleaning the data. Getting the data in the right places is really critical now.
AI Announcer 0:49
Welcome to the conversations on Applied AI podcast where Justin Grammens and the team at emerging technologies North talk with experts in the fields of artificial intelligence and deep learning. In each episode, we cut through the hype and dive into how these technologies are being applied to real world problems today. We hope that you find this episode educational and applicable to your industry and connect with us to learn more about our organization at applied ai.mn. Enjoy.
Justin Grammens 1:20
Welcome everyone to the conversations on applied AI Podcast. Today we're talking with Fraser Gray-Smith. Fraser is a business strategy professional that specializes in using analytics to drive value in organizations. He takes advantage of his unique mix of technical skills and business acumen to uncover valuable insights, develop business cases, create management buy in and manage the execution of projects. Outside of work, Fraser likes to learn about leadership and politics. He also enjoys experimenting with emerging technologies in the finance and machine learning fields, which is exactly what we like to hear hat conversations on applied AI. So this will be really, really interesting conversation. Fraser, thank you for being on the podcast today.
Fraser Gray-Smith 1:57
Thank you so much for having me, Justin..
Justin Grammens 1:58
Well, you know, I gave a little bit of background with regards to you know what you're doing today professionally, maybe you could shed a little bit of insight in terms of, you know, maybe your upbringing, and then you could maybe even start during college or whatever. But you know, what, how was your professional life been?
Fraser Gray-Smith 2:11
Yeah, it's funny, I actually I kind of fell into tech by accident a little bit. My formal background was was in finance, I studied the finances and undergraduate and I always kind of liked the numeracy side, like I liked the the numbers side. But formerly, I studied to be an accountant of all things, went through the courses, a data, a couple of internships, and kind of accounting and banking and decided that it wasn't for me not to say that, you know, for those accountants out there, I'm sure you're very happy, but wasn't for me, and then went back to school and did my Master's in international business, which was really just kind of an excuse to travel for a year and get a get a degree. And so once I kind of finally figured out that I needed to get a job and be a productive, productive member of society, I found a role at Bell Canada, which was Canada's largest telecom company, it was a mixture of strategy and analytics. And the way I sold myself was, hey, I'm, I can kind of do this strategy thing about a finance background. And I'm really enthusiastic about analytics, because I was I, you know, I discovered that I liked the analytic side, and we liked the number side. And from there it was, it was just a rocketship taking on more and more senior analytics roles. And I discovered that I loved the very kind of hard technical side as well. I'm somewhat of an impatient guy by nature. So I hated the fact that I had to rely upon some of the business intelligence teams and the data scientists in order to get results. And then it would be I would say, Okay, well, what if we looked at the data this way instead? And it would be the answer would be okay, we'll leave the other two weeks to do the, to do the analysis. So I taught myself how to code and right kind of at the time, it was no SQL and Python, really, really loved that took on increasingly more senior roles, doing a little bit of machine learning, and found that that was fascinating to me. And then, more recently, I have gone back to school part time to do my master's in computer science. And I moved away from ballet to about takedown and a couple of other roles. And I'm currently at Slalom Consulting. And we're at we're a global consulting firm, we're focused on strategy and technology, strategy, technology and business transformation. And what that really means kind of practical use, I find ways to apply all my analytic skills in order to solve business problems, which is really what excites me. And that's what gets me up in the morning.
Justin Grammens 4:27
That's awesome Fraser. Yeah, I mean, so many things to unpack here with regards to your background, and you know, who you are as a person, like you saying, you're, you're an impatient guy, I think I find some of that too. It's like, jeez, if I can just get in there and do it myself. Why don't I just learn the tools and sort of do it? And in your current role, it's, you know, and I even said it during the interview, it's or during the intro, it's really you're dealing like with emerging technologies, right. And for our listeners, actually, the applied AI podcast sort of lives under a nonprofit that we have formed here in Minnesota called Emerging Technologies north and so being able to to sort of bring all those aspects together, whether it become, you know, your background in the finance side, the work you're doing in, you know, initially analytics now getting into machine learning, being able to sort of apply that to business really is, is a great, I guess, use case of how all these technologies can sort of come together and make business transformation happen. What are what are some of the things that you're seeing, I guess, you know, like today with regards to how companies are using these, these these technologies to change how they do business?
Fraser Gray-Smith 5:25
Yeah, really good question. I think the number one are a couple of trends, I'll say number one, but there's a couple of different trends. The first is that machine learning and AI has really become a little bit I'm gonna save democratized, there are so many tools out there today, that didn't exist even five years ago, that allow anybody to build AI driven models. And these are point and click Tools, if you have basic computer skills, if you're a capable of you know, operating a Facebook account, then you can probably use these tools, they're all very, very simple and straightforward. And they allow people to really take advantage of AI models. Now, they'll never be as advanced as you know, a dedicated data scientist who can go in and do a ton of more advanced modeling and, and really tune the models in order to make sure that they're, they're perfectly aligned for that specific business problem. But what it does is it allows everybody at a company to go ahead and and build AI solutions to problems, which I think is fantastic. Alongside that. So in addition, the kind of flipside of the the democratization of AI, and the fact that everybody can build it, is the fact that it gets misused a lot. The idea that AI can solve every single problem is MC or AI is appropriate for solving every problem is maybe not as true. I think that there are certain problems where AI is very good at solving and very appropriate for solving those sorts of problems. But AI gets kind of used as this hammer and kind of wax, you know, you're playing Whack a Mole with all your problems and just trying to use AI to solve everything. I think that that that happens a lot to where especially I see people who are maybe less familiar with what AI actually is or what it does. And they say, oh, we'll just throw AI at it. And it kind of becomes this catch all solution where maybe it's not appropriate. So good and bad that I'm seeing at F companies.
Justin Grammens 7:22
Interesting, interesting. My background has been a lot in the internet of things. And so you get it, you kind of go through this hype cycle, right, that Gartner Hype Cycle, which I'm sure you know about that. And, and everyone's like, Yeah, this is the answer to everything. And then kind of realization hits. And you get into this sort of trough of disillusionment, where you just kind of like, okay, what are the business applications? What are the true problems trying to solve, and then you finally get out to the plateau of productivity. And so I like really like to talk to people a lot about that, because they feel like, feel like AI is different. And machine learning is different than IoT, obviously. And when I'm doing these communities conversations, and having these interviews with people, I do feel like there's some real, one of one of them of the people that I interviewed, you know, a number of months back said, you know, with machine learning, though there is there's rubber meeting the road, right, there is actually no cheese, they can calculate the amount of time savings that companies are saving, because they are, you know, using these new technologies, where some of the other stuff, you know, whether it be blockchain or whether it be IoT, or additive manufacturing, some of the other emerging technologies is kind of kind of tough to really say, Okay, this is this is where it's actually happening. So, I mean, are you as you're consulting with, with companies, what what sort of like industries are you working in? Are you typically working in, in in finance and trying to maybe, speed up, I guess, take away some of the mundane tasks that companies are doing in this space
Fraser Gray-Smith 8:35
general comments on a couple of different industries, I find the tech companies that I work with, they've got a pretty good handle on what is appropriate for what is appropriate use cases for AI and what is not, they're sort of one step ahead. In terms of adoption, which sort of makes sense nature of the of the industry, I find that finance is really starting to come around, especially in the banking realm. I think that there's there's a lot of regulations that are starting to catch up. And, you know, folks who are listening who probably understand, you know, banking very well, it's a very regulatory driven industry, there's a lot of kind of red tape in terms of what you can do and what you can't do, and how the different ways that you can use data. But I think a lot of the regulations are catching up, which is enabling more financial institutions in order to take advantage of AI. One really good example is there is a Canadian slash US Bank, it's a TD Bank. And a couple of years ago, they acquired a startup called layer six, which is a very kind of AI focused startup. And they're really kind of shaking up the space. They're doing a number of things in the banking realm, which have never been done before. challenging some of the traditional credit models and some of the traditional risk models using AI that you know, have haven't been done in the past. So that's a really good example of that happening within financial institutions. And then the other place I'm seeing AI be used a lot is in any sort of retail practice in order to enhance customer experience, I think one of the things that that sometimes gets missed is that customer experience can be really, really important, especially today, when as a consumer, you have the option to either buy something in a store or purchase it online. Unless there's a compelling customer experience, or reason that we would make you want to go into the store, you're tend to lean online, it's much it's more convenient to shop around and find the best price. So I think retail companies are starting to come to that realization and say, Okay, well, how do we make? How do we use AI to do personalization in store? Or how do we use AI to to drive a better customer experience, and make people want to go into the stores and get? So I think that's, that's an interesting, a couple of interesting comments about a few different industries.
Justin Grammens 10:51
Yeah, that's fascinating. There's banking going on. And then where I've seen a lot of, at least a number of companies here in the Twin Cities attacking this, this mortgage industry, there's, there's so much paper, and so many humans need to read over so many documents when you go to close on a property. And so I know a couple of different companies here that are attacking that space, in particular, that can just essentially, but it's an interesting problem, you know, there's a lot of essentially, you know, computer vision, some some, you know, image to text stuff that needs to needs to be done on these on these documents. But, boy, there's a huge amount of time savings and essentially, you know, human capital that can be freed up to do other things. And that's, that's what I like to tell people is like, we're not gonna be putting people out of jobs, they're just not going to be doing more higher level things, right? Yeah, absolutely.
Fraser Gray-Smith 11:35
You don't need somebody necessarily to read a document and physically key in the words, you know, the person's name and email and phone number from that document into a system, you can have the system do that, and the person can actually focus on what is the content of this mortgage application? And does this make sense to extend credit to this person? It's much more kind of thinking tasks rather than the manual, the plugging numbers into a spreadsheet type of tasks?
Justin Grammens 12:07
Absolutely. So we've been tossing around the term AI a lot in this in this in this conversation, I do like to ask people like, how do you define it, I guess? And if somebody says, Well, what do you do in your day job? And you say, Oh, I work on AI solutions, or whatever. They're like, well, what's AI? Do you have a ladder, no succinct version, or anything you typically kind of try and describe a is
Fraser Gray-Smith 12:26
my favorite definition that I like to use is AI is enabling decision making using applied math at scale. So the three parts of that, like to break it down, like enabling decision making, really AI should, I don't like using technology for the sake of using technology, I like to use technology in order to actually solve a problem. And a lot of the time that's, you know, with applied with artificial intelligence, you're actually making a decision. That's the way you're that we're actually using technology. The second part is kind of the Applied Math at scale. The reason I like that is because a lot of these techniques that we're talking about artificial intelligence, a lot of these, the actual mathematics behind it, and it has existed for a long time, things like, you know, logistic and linear regressions, okay, well, they've been around for at least 100 years, if you go down kind of the path of using AI for marketing science, as an example, that I have read talks about Markov chains, when Markov chains were made up in the 50s. It's just that we now have the technology in order to do those calculations at a scale that we haven't seen in the past. And we have the ability to do the calculations and give a result instantaneously. And that's why that's kind of the game changer here. For AI, it's that there's not that, you know, we're coming up with entirely new techniques, we're using the techniques that we have, but we're doing it at a scale, a new scale, and faster than we've ever done before. That's what's helping us make better decisions.
Justin Grammens 13:54
I love it. That's, that's great. I like a couple of keywords that are in there. One is the decision piece. And and so kind of like looking at it more from the what problem you're trying to solve more humanistic thing. And then I majored in applied math myself, actually. And I jokingly tell people that you know, I got sick of solving for x like, well, I don't want to do traditional math, like just solving equations, like I want the applications of it. So applied math and physics were sort of like my jam when I was in my in my undergrad, because I'm like, I really want to figure out things in the physical world. So that whole application of it, and then I, and then and then tying it into the whole scale, because you're right, a lot of these neural networks, a lot of these techniques have been there. We've just now now we were sort of like a we're getting the compute power to be able to do a lot of these interesting things. But then also be maybe you can speak to this a little bit. It's awesome. These tools are out there for for for people, but there's still a lot of work around cleaning data, actually getting data, you know, I guess formulating in the right way that you can actually, you know, run models over it, or is that where you're sort of seeing now, the bigger challenges because you're right, it seems like there's so many tools out there where you can just sort of throw things into TensorFlow pytorch, all these other ones, and even things that you write online, people can just drag and drop data. And but man getting the right data, you know, set up the right way and having it cleaned and all that type of type of stuff. That to me seems like a big challenge. But you know, I don't know, are you seeing that as well?
Fraser Gray-Smith 15:15
Yeah, absolutely. I think the data cleaning part of any data science project usually takes about 60 to 80% of the work, and only about 20 to 40% of the work is actually building models. When we talk about data science projects, or at least that's that's been my my experience. In addition to kind of the the nuts and bolts of data cleaning, I'd say the other task that needs to be done more today than ever is building data pipelines. I think that there's a proliferation of tools, there's a lot of different tools out there that companies use, that are generating data, and they want to connect all they wanted to connect all of these data points together, they want to bring data from lots of different sources into one place into one model. And I think that's the right thing to do. I think generally speaking, the more data that you can put into, especially some something like a neural net model, or some of these larger, these more complex techniques, I think, generally that does improve the quality of model. But because you've got data residing in all of these silos, one of the things that has become an integral part of the AI process, is this data engineering, you know, how do we take data from these 10 different systems and put it into one place and connect it all together in a way that makes sense? I think that that's that in addition to cleaning the data, getting the data in the right places is really critical now?
Justin Grammens 16:42
Yeah, for sure, for sure. Well, what's, what's a day in the life of a person in your role?
Fraser Gray-Smith 16:47
Whenever you ask a consultant, you know, what is the day in your life be like, a lot of consultants like to say, well, it's very different, because, you know, we're always doing different things. And I think that's partially true. I think that consultant by nature varies based off of the client based off of the industry based off of the problem you're looking to solve. However, there's kind of three main things that I do as per barrel. The first is I asked a lot of questions. The second is I build things. And then the third thing is I talk about having built things. And that's kind of the most general description I can get. So asking a lot of questions. I think the first part of any, any good project is you got to sit down with your stakeholders. And sometimes you got to ask almost ad nauseam, you know, as many questions as you can, what are you trying to solve? Why you're trying to solve them? What's your real goal here? So a lot of folks will say, Well, my, my, my, with the problem, I'm trying to solve this, I'm wondering marketing attribution model, or I want to turn model or once you know, whatever, okay, but like, what is the problem? You're like, what is what are you actually trying to do? Are you trying to increase your revenue? Are you trying to decrease your costs? Are you trying to, like, what's your business objective, so you got to spend a lot of time can probing for those deeper answers, in order to really understand what's the size and scope of the problem. The second part is building as a built in stuff. That includes everything from data cleaning, doing the data engineering, as I mentioned, piping all of the data in one place, building models, tuning models, making sure that they are performing at scale, deploying them, that falls all under the kind of a building category. And then the third part is communicating. And that's really just going to our stakeholders and saying, we have done X Y Zed, here's how it works. Here's when it works, here's when it doesn't work as well, the limitations of the model, the assumptions, here's what went into it and and how you can use it in order to functionally do something different or better with your business. And I think the somewhat unique part of consulting is that in many organizations, if you're in house and I say this, having been an analyst and a data scientist in house, you may be restricted to one part of that cycle, you may, you might have a business analyst who asks the questions, the data scientists who build stuff, and then some sort of manager or somebody who communicates what was built. The unique part of consulting is you are the person who's doing all three. So you get to see a project end to end, you get to see a little bit. So a little bit of everything, which is, I think it's a lot of fun gives a good breadth of experience.
Justin Grammens 19:23
That is great. As you were sort of talking through this, I was writing some notes and this might actually be just good career advice for for everybody. Right? I think the ask a lot of questions kind of shows people that you kind of need to be a lifelong learner. I think as you build things, as you said, I view it as take action or take initiative, right? So no matter where you are in your career, you should jump in and take initiative and, and then you know, talk about the things that you built. You touched on it very well. It's just you gotta be a good communicator. So you know it no matter where you are in your career, if you're just starting out, or you're a seasoned professional, I guess you kind of want to be a lifelong learner take initiative and also communicate very well. This leads into my next question, I guess is, you know what, how would you maybe coach somebody or like mentor somebody, I guess it's this maybe just coming out of school and getting into the field, like, what what are some interesting things that you've seen, whether it be groups or books or, or other resources, I guess that maybe you suggest you point people to,
Fraser Gray-Smith 20:17
as I started to get kind of deeper into data science, I had a number of mentors who I spoke to, and I was a couple years out of school. But very similarly, I didn't have the kind of the technical background. So I had to start from from step zero. And I think that for somebody who's looking to break into AI, you've got to understand at least a little bit of programming and a little bit about the math in order to really appreciate what's going on. Now, I'm not saying you have to go back to school and get a computer science degree, although I am doing that. And I'm not saying you have to go back and understand the math. But I think from a resources perspective, if you want to get started on on programming, Harvard posts, their course of CS 50, it's free to take a you can just Google CS FDN, and lectures are on YouTube, and they post a website where you can go in and actually do do the programming challenges. It's difficult, it's rewarding, it's interesting, it's taught in a way that's very accessible. But it also makes you think, so I highly recommend anybody who's getting started with with AI. And if you don't have any programming background, and start there, from a math perspective, and I know, folks that can be a little bit hesitant when it comes to the math, MIT has a series of lectures, it's true, it's in what's called OpenCourseWare. And they have a number of undergrad courses in math, they, again, they post the lectures, it's videos, watch it and understand what's going on. It's very, very accessible. So I do recommend that if you just want to kind of understand some of the concepts and so when people are talking about, you know, something related to matrices, you know, okay, well, that fits under linear algebra, and I can kind of understand what they're talking about. Once you've got a good understanding of CFD and or you're gonna understand the programming and the map. The final part I would say is, Andrew Ng has an amazing course I'm sure you've seen it before or heard of it before. It's it's a machine learning, it's taught through Stanford, it's online, again, free course, anybody can take it, it is top notch, fantastic. If you can, if you do those things you will probably be ahead of, I'm gonna say 60 to 70% of the data science applicants, I see. It's those three things, if you have a good grounding in those three things. That's amazing. And then the only the other advice from a career perspective that really helps me is build stuff that's interesting to you. So throw, you know, as you're taking these courses, don't just take the courses, take them and do something with them. And the thing that if you want to get into AI, the thing that could do that is build something in the topic that you're interested in. So for example, I'm Canadian, so stereotypically, I like hockey, which is actually true, I do like hockey. And so as I was learning some of these data science techniques, I built things that were related to hockey, I built models that predicted who was going to win the next state who was gonna score the next game more, you know, what the standings were going to be at the end of the year, because the topic was interesting to me, that helped me apply what I was learning and in new ways, and interesting ways, kind of stretched my skills a little bit. And obviously, it doesn't have to be hockey, if you're interested in finance, or politics, or healthcare, or you know, any one of 100, movies, whatever, whatever you want, you can find something, some way to apply the techniques that you're learning, in a way in a topic that's interesting. And that'll help you that will help you understand the topic a lot more.
Justin Grammens 23:44
That is great. Thank you for all the resources and we we always have liner notes for for our podcasts so people will can be able to I will post all this information in text form and people will be able to click off to this Harvard CS 50 and the MIT OpenCourseWare. Andrew Ng, you're right, is he's like a legend, right in this whole space, since he was one of the first people to start doing machine learning at Google and stuff. And then also created a created this this whole course, which blows my mind. Like I mean, just think back like, like if we didn't have the internet, like how would people pick this up? Like how would people like learn this? It literally would just be printed books, and I guess, you know, word of mouth and phone conversations and whatever it is, right? I mean, just it would be a lot. It would proliferate a lot more slowly. I guess. It's amazing that there's so many things online. Yeah, no, you're
Fraser Gray-Smith 24:33
Justin Grammens 24:34
I also really love you talked about doing something that you enjoy, because that's really is what's going to, I think drive you to the finish line. If it's something that feels like it's a chore, it's like, oh my gosh, I need to do this math thing again tonight. It's not going to get you to where to to finish. It's not going to be as rewarding and it's also not going to get you to where you want to go. So yeah, if you can do it through something you enjoy. It's going to feel much more like a hobby than something then something that's hard to do. Are you doing any reading these days? is one of the questions I do like to people is like, do they do you have a favorite book? During the intro here? We talked about leadership and politics. I mean, have you are you are you exploring some of those spaces and reading some interesting things in that area just outside of this tech
Fraser Gray-Smith 25:12
outside of the the AI realm, I finished a book, it's called so good, they can't ignore you. It's by Cal Newport, it's a bit of a different kind of leadership slash career philosophy is central idea is that the career advice of follow your passion is is actually terrible advice. And I know that sounds, it's a bit shocking, but the way he talks about it is we as humans, we start out not feeling good things, any new skill, you want to pick up via AI, or if it's programming, or if it's, you know, you want to pick up the guitar or start to learn to play the piano, you're going to start poor, that's who we are, nobody comes, nobody is born, knowing how to do any of these things. But as you start to practice, and as you with constant with consistent practice, you become better. And it's the fact that it's the act of becoming better at something that makes you passionate about it. So the passion doesn't, doesn't come in advance, it doesn't come when you are deciding what to do. In fact, if you decide what you want to do, you can become better at it, and then you become passionate about it. And that further incentivizes you to do more of it. And so it becomes this sort of virtuous cycle where you become good at something that feeds your interest and your passion for that subject. And that which further incentivizes you to get even better at it and keep practicing, which even further makes you more passionate about it. So it's a bit of a different paradigm shift. But it talks about, like, Hey, you don't pick something because you think you might be Don't, don't do something. Because if you think you might be vaguely interested in it as a career, pick something that you know, you think you can, you can become good at, become good at that thing. And the passion will follow afterwards, and you will become interested in over time. So I thought it was really, really interesting. It's a different take on kind of careers, but I liked
Justin Grammens 27:07
it. Great. So yeah, so good. They can't ignore you, I'll be sure to put put a link off to it. I know, I've heard that, well, at least Cal Newports written a couple of different books, I think as face just just around professional development and and I guess bettering yourself in a number different ways. So he's a great author, it's cool. It's sort of like reframing, or just a sort of a, I guess, a mind flip with regards to sort of like what comes first. So that's awesome. And, you know, I guess what I found too is is, you know, the more outside reading you do outside of tech, outside of like, you don't need to be sitting here reading all about AI and machine learning the entire time. Like, I feel like actually, it's good to pop your head up, read some of these other books that are just completely different and actually makes you a better a better professional, better technologist.
Fraser Gray-Smith 27:49
Quick aside, one of the most interesting courses that I use more now than, than almost any other, I'm not gonna say any upper, but almost any other and my my undergraduate is I needed a writing credits. When I was that it was during my undergraduate, the only course that was offered that fit my time schedule that was a writing credit, was ancient philosophy. And it was probably one of the most difficult courses I've ever taken. And it was, you know, the reading was very dry reading, probably 2021. At the time, as you know, saying, Well, I'm just reading about these dry philosophers, because I'm this writing credit. And so I'm, I'm going to do this course I'm going to get through it. I think I passed 65 or something like that it was not a very good mark. But the things that I learned through studying philosophy of all things, like how to make a reason and things like how to make a reasoned argument and how to structure arguments, or how to dissect somebody else's argument, and what makes a strong argument versus a weak argument, or you know how to examine different points of view and hold different viewpoints in your mind without agreeing or disagreeing with them, just acknowledging them as different viewpoints. I use that skill every single day, when I'm in front of a client, and they'll say, Well, I think this okay, well, let's, let's dissect the logic behind that. Why do you think that and what is what is driving that that thesis, that hypothesis? So it's very, very outside of tech, you think? What does ancient philosophy ever have to do with, you know, technology or AI, but it actually did end up helping me at the end of the day.
Justin Grammens 29:33
That's awesome. I was thinking about there's another author, and he mentioned, named Benjamin Hardy. And he's a behavior scientist. And just like, why do we behave the way that we do? That has been really fascinating for me to read a number of different books in that space, and understand Oh, wow, I don't Yeah, I guess I guess I do do certain things a certain way. You know, there's always room for improvement, or ways to change things and sort of like it's the human mind. Mind is always adapting to new things. So it's a fun space to sort of read and explore are in Sure. And then of course, tie it back into, you know, business use cases and machine learning and AI and and actually how the technology applies. Are you guys hiring at Solem? I'm assuming
Fraser Gray-Smith 30:10
we are we're very aggressively hiring. I'm part of the global team. So I get to see compliance around the world. But we also have very, very many local offices that are hiring for all sorts of tech skills. So I cannot recommend enough. They're not paying me to say this. I'm saying it because I honestly believe that no, I actually do. They're, they're a great company. I can't recommend it.
Justin Grammens 30:31
That's awesome. So yeah, I will get a link to your careers page. And I'll post that here as well. So Frasier, this has been, it's been awesome. How do people reach out to you? Is it easiest, just to find you on LinkedIn, Twitter, Instagram, where
Fraser Gray-Smith 30:43
do you please find me on LinkedIn, I have a pretty public presence there. I don't think I've ever said no to a LinkedIn invitation. And, you know, schedules permitting, I will almost always say yes to coffee, virtual coffee chats, or, you know, if you're local to Toronto, you know, happy to do it in person as well. I love chatting about this stuff. And I love helping other folks discover a little bit more about the space. So more than happy if anybody listening wants to reach out and happy to have those conversations.
Justin Grammens 31:12
That's awesome. Fraser, well, thank you for all the work that you do in the field here. I always have interesting conversations with people and just just really fun for me just to sort of explore what people are doing in the space and I get a chance to learn and all of our listeners get a chance to learn from from great technologists like us. So thank you for being on the program today. I appreciate your time.
Fraser Gray-Smith 31:31
Thanks so much for having me, Justin. I appreciate it.
AI Announcer 31:34
You've listened to another episode of the conversations on applied AI podcast. We hope you are eager to learn more about applying artificial intelligence and deep learning within your organization. You can visit us at applied ai.mn To keep up to date on our events and connect with our amazing community. Please don't hesitate to reach out to Justin at applied ai.mn If you are interested in participating in a future episode. Thank you for listening