Tito Tech Episode 3: Working in Big Tech, Talking Big Data and Taking Big Risks with Janet Uy
VO: Created by UBX. Mario: Hi everyone, I'm George Clooney. Oh, no I'm not. I know it's not obvious. But Tom Cruise? I’m Coco Domingo. Welcome, welcome, welcome. I am Mario Domingo, the Global Chief Technology Officer at UBX.
And this is Tito Tech. I am your Tito for all things technology. Today is a special day because I have a guest. It's going to be a series of other guests, but today is very special because here, we're going to have a very good friend of mine, Janet Uy, who's the Senior Cloud Solutions Architect at Microsoft Singapore. Welcome, Janet.
Janet: Hey, Tito Mario. Happy to be here. Thanks for having me in your Tito Podcast show. Mario: Nice. Yes, yes it is. For our Tito Tech listeners, Janet and I go way back. We're familiar with each other's work. She used to be a management trainee. She left to go do her studies in Singapore where she was able to secure her PhD in engineering. She's moved on, she's built a career in technology, specializing in data and artificial intelligence. Then, so maybe Janet, you can reflect a little bit on like,
your experience in the early days, what's it like, and what made you go to Singapore? Well, I remember, so as Mario mentioned, so I was a management trainee at Globe Telecom, and he was my first boss. So I'm not sure whether you chose me, or it was just serendipity, and you became my manager. That was kind of the first exposure to tech and I got really inspired and six months into the role I got the opportunity to be offered a scholarship at Singapore. And I remember I was like, so nervous. I walked to your room and was like, Mario, I think I'm
gonna resign. He was like, why? Right? So because I was in a dilemma, like should I go or should I not go because I really enjoyed working with you in the first, I think four months or three months, right, like before we rotate and I came to your office and it's like, I got offered a scholarship, but I'm not sure whether I should go because I felt like I was in a very nice position already. It's very hard to get into Globe. So, but then I think you talk to me, removing the boss hat, but just ask me the question that do you think it's the right thing to do? Would it help you in the future? You're young, go and, you know, go and do what you want and what you think is right for you. So, there. She imagined most of that, okay. No, I'm just kidding. So yeah, I kind of remember,
I remember this high potential, high potential young girl coming to… into the telecoms field, but you had the opportunity to get a scholarship. That was a good thing. And then what happened? What happened next? Yeah, and so fast forward. So after school, I worked. And then, I don't know, again, maybe it's
serendipity. I received a message from Mario. It's like, what are you doing? Do you want to come back and join me again? Like where? What am I supposed to do? In a startup. So because I think before I left Globe, you said that go and catch your dreams but then somewhere down the road, maybe we'll meet again and we get to work with each other again. And then that message came and I was like, huh, is that a calling? Is that a sign? So… and that's when I came back and joined Mario at DANATEQ. Well, DANATEQ was a, basically a software high speed database powered by AI technologies and it was a very successful venture for me. Janet was along with the ride
for that. She did very well in her rollover then and then what happened next? You went to? lloopp. There you go. And so we went into the second startup, also very successful. But Janet didn't stop at just the technology startup company. She's moved on to bigger tech.
So as Mario mentioned, after that, I joined Oracle. And after Oracle, Microsoft. So pretty much throughout those years, I was in the data science, AI, ML space. So you've been through the gamut. You've been through the startups we've been,
not to mention a highly confidential partnership, not too far from here, where we looked at insights and behavior and we're always pushing the envelope on combination between quantitative models to what it means on a qualitative perspective. What do you see in terms of how the trends and evolutions have played out over the years? Do you feel that some of these technologies that we dabbled in in the areas of deep learning and machine learning, do you think it's delivered the promise to the enterprise? Or do you think we're still relatively in, early stage or early age of that trend? I think especially on the AI space, right, it has evolved and improved tremendously in the past two years, I would say. And mainly driven by, I think, the data availability, the data collection. So I remember in the early days when we talk about building the models, the main constraint is about data, right? So we have to design how to collect data, how to manage the data, how to clean the data. Do we have enough data? That was the big question that we have.
And even training the models, like we get hit by computational limits and all these, and it gets very expensive to train models, right? And so the value that we get out of it, I feel that before it was very difficult to justify the investments in AI. But now, given data is almost like the new blood, it's everywhere, and with IoT devices coming out, I think pretty much, on a day-to-day we have data everywhere. And collecting that is a lot easier, but of course with additional challenges on, as you mentioned, data privacy and all. But the technology has evolved a lot and the compute power is just amazing at this point in time, the models have all improved. And so with AI, I think it has become, we can call it like
it has become more like democratized, right, in a way that it's now infused in our day to day, not just enterprises. I think used to be enterprises were the ones who could invest and spend money on AI and building their AI projects, but now it's pretty much everywhere. So I would think that in terms of value, that more and more people are now able to benefit from the technology.
Yeah, that's right. Even in startups today, there's accessibility to AI that was not apparent or available years ago. So you're well-traveled, you see the global north, you see Asia-Pac, right? I've always had a cynical view about the inequity of data. When I say the inequity of data, those who have and those who have not, right?
Say for example, in the case of the Philippines, where some, not so much the big guys, not the top one to 2% of the large enterprises in the Philippines, but the next year, where many of the leadership is struggling to find the business case for digital data, machine-generated data, its storage and their old… for sure they want they AI. For sure they want predictive capabilities, but they don't want the price of cleaning the data, storing the data. And can you comment on that? Yeah, I think, like, that has improved as well, right, like with the emergence of foundational models, right. So you have the big corporations training these foundation models like, I don't know if I can call it, you know, open AI, like for example, has invested a lot on that. And so companies, who do not have that much data,
can actually build on top of these foundational models. So companies really would have to just focus on what data is important to their use cases, to the business, and leave the rest to the foundational models to augment whatever is lacking. That's another way of saying that in the old days, we have to find an algorithm, we need to stack it up, we need to layer it in, in order to get to a starting model by which we can start creating predictive stuff from. What you're kind of sharing is that in today's state of things, those foundational models are in place, basically shortening the time to insight, shortening the time to prediction and output, right? Exactly, yeah. So like right now, I think we can confidently say there are about thousands
of foundational models that are available out there. And then you can just connect or add your data or layer your data on top of it to arrive at a usable use case. So a use case that can bring value to the company on day one. And it's very easy to build. Like with generative AI,
I think building applications. There are tools available that companies can leverage. And there you go, you have your application in a few hours, sometimes even minutes. So who is your, on the client side, who's the advocate? Is it the CTO, is it the CIO, is it the tech guys? Who's the advocate for generative AI? I would say it still goes that, you know, the advocacy comes from top down, right? So from the C level, but then for these to be useful to the rest of the organization, so you need a bottom-up approach where you need the users to really test it and bring the use cases out, right? And then these two will meet halfway, yeah. Where do the majority of the use cases come from? Is it from the business side of things? Business side. Yes, that's what we're seeing, right? So from the everyday users like mainly on, for example, on the productivity, on the automation, operational improvement, I think for the customer support in the end-to-end operations, right? So this is where we see the biggest value. Right. So just like anything technology, just like anything technology, it should be the business
that should be spearheading the initiative. Don't delegate this to your technical teams, you should be the one to initiate and advocate. You can't do this if you expect it to be served to you. You have to invest time and effort in studying this thing. So if I could segue onto financial industry, what are you seeing in the financial industry, the banks, in the general Asia-Pacific area, what kind of trends in data, AI, or analytics do you see? I think the fintech or financial industry is probably one of the most advanced in terms of adopting analytics and AI, right? So I think first they have the capital, right? Like the resources. And second, I think, you know, going back to what we discussed earlier is that, the availability of data. The financial companies would be sitting on a gold mine, so they have a massive amount of data for customers, transactions, payments, and with all the alliances and partnerships happening, fintechs with retail, for instance, fintech with all the other industries, the data that they're sitting on is even larger. So I think that's… With that,
it becomes a very nice playground to test out generative AI and all the other AI applications that we are seeing. So, massive, I would say, improvement or trend in Fintech. So what's next? So I get asked this all the time in my talks, right? Mario, are we going to find singularity? Is there going to be sentience? Is, you know, is generative AI going to be able to emulate social intelligence? What's going to be next for AI? I think right now we are just scratching the surface of AI. There's still a lot more to do with generative AI. Like we are seeing use cases around productivity, but then there's also healthcare, there's education, like we talk about finance, but retail, customer experience, customer service, like there's so much more, right, like in the horizon. So I think like even with, like the improvement of the models last time when we talk about deep learning, we are looking at NLP for instance, natural language processing, we're talking about image, video separately, but now we have multimodal models, right? So we haven't kind of exhausted that space yet. So I think we still have a long way to go when it comes to extracting all the use cases and the value out of the generative AI and then maybe in the future we have the more general AI.
So I was reading a McKinsey report on general AI and it's, very gloomy kind of outlook, right? Is it mostly to generate clicks and hype? What do you think? I think so. I think for us who are in this space, we have a role to play in terms of promoting responsible use of AI. And we are all advocates to make sure that we create that awareness on how to use and train models responsibly and properly. So I think that it's not, like in everything, there's always a good side and a bad side, but the value that we can generate out of AI, it's really, I would say, unbounded at this point, right? So we can keep leveraging the technology to do good and help create value, help users maybe even augment the skills, and for us to focus more on what we can do best as humans. And then we can
then leverage AI again to mitigate any risk that is out there. So I think if we all work together and work towards responsible AI, then I don't think it would be that gloomy. Good. Brilliant. That was great. So I have a couple more questions,
then we'll go to the call-in question. Do you still play tennis? Yes. Janet is a star tennis player at De La Salle University. She's from La Salle. She's from De La Salle University. She was varsity over, in De La Salle in 2015, something like that. Yes, that's how young I am. But how does your work and how does the sport? Well, I come from, I play football, I used to play football a lot, right? And so there's, there seems to be a link between the way I, my, the way I manage myself and the way I react to things that comes from that sports thing. Does that have the same impact to you?
Yeah, I think it's the… number one, it's the discipline, I would say, right? Like making sure that you're well-prepared, that you practice really well, you're all set for a game, that's one. But I think the other one is the mental toughness, right? Like, it's like never say die until the game is over, right? So, and this is the same thing in our day-to-day, in businesses, right? Like every day you feel that you're almost at the verge of giving up, but then no, no, no, I have to wait for the game to end and then, yeah. So I think that's really the correlation between the sports and work. Correlation, folks. Correlation.
Correlation. We now go to the final segment where we've got a question from our listeners. In this case, it's from Kate. She's a data lead at the, she's a data lead from a FinTech company. She goes,
Oh, look at this question. Hi, Janet. What do you do for fun? For fun. Okay. What do I do for fun? Nothing? You sleep, you read books. Oh, you read books though. Do you still read books or, no? I read books. So what do I do for fun? Janet works all day long until all night long. I'm a night owl. Right. So like, I remember I usually sleep very, very late at night,
sometimes until 4 a.m. and 5 a.m. and I will send all my deadlines and work to Mario by that time and he'll wake up and then check my work and then while I'm sleeping, he'll wake up again and then done. I got all the approvals. But what I do for fun, I think I really love learning new things. So even though, right, like I now manage teams, but I try to be updated with technology. So I still do online learnings. I still study. You're in school today? Yes, I'm doing my EMBA.
You're getting your MBA? Yes. In? NTU. At NTU. Nanyang Technological University.
And I heard you have a project and you have some kind of a sponsor. Yes, it's called, our sponsor is UBX. Yeah, yeah, the group's– Yes, and they'll be infusing AI and technology into our project. That's what we do for fun. Yes, yes, very good. That's a wrap for today. Thank you, Janet,
for joining me. It was a lot of fun having this conversation. We picked up a lot of stuff along the way. I hope to have you again one of these days. Thank you so much for having me here. Yes, and it was a lot of fun. And yeah, like next time we can do this in Singapore, along the river.
Yes. Don't forget to click the links, turn on your notifications, make sure the bell rings, and subscribe. Welcome to the show.
Hindi ko alam. Muntik lang. Ano ba? Ganun. Glad to be here. Something like that. So parang– Please forgive her. She doesn’t know what she’s doing. Yeah. It’s alright. We will do it again.
2024-10-31 08:42
