Our discussion today centres on the profound transformations occurring within the realm of product management, propelled by advancements in artificial intelligence. I had the distinct privilege of conversing with Brian Yan Muk, a product manager at Apply Digital, whose extensive experience encompasses leadership roles at renowned organizations such as AT&T, PGE, Moderna, and the NFL.

Throughout our dialogue, we delved into the intricacies of leadership in both small and large organizational contexts, emphasizing the importance of empathy in decision-making processes. Furthermore, we explored the implications of AI on product development and the evolving expectations of leadership in today's digital landscape.

Brian's insights not only illuminate the path forward for aspiring leaders but also underscore the necessity of adaptability and continuous learning in an ever-changing environment.

Note: The views shared in this episode are Brian's own and don't represent the views of Apply Digital.

Takeaways:

• Brian Yan Muk has led product teams in prestigious organizations, demonstrating significant leadership skills.
• The essence of effective leadership lies in empathy and understanding team dynamics and individual contributions.
• AI is reshaping business processes by automating tasks, thus requiring leaders to adapt their methods accordingly.
• Successful product development hinges on clear communication and stakeholder involvement throughout the process.
• Effective leadership necessitates decisiveness, even if it involves making unpopular decisions for the greater good.
• Continuous curiosity and experimentation are crucial for adapting to the rapidly changing technological landscape.

Links referenced in this episode:

• www.applydigital.com
• www.sfu.ca
• www.enactus.org
• www.linkedin.com/in/brianyanmuk

Today's guest is Brian Yanmuk, product manager at AppliedIgital.

He's led product teams for some of the biggest names in the world at&T, PGE Moderna and even the NFL, scaling a team of over 80 product managers, engineers and designers across North America and Latin America. Before any of that, he was coaching case competitions at sfu, a passion that never really left him.

Today we get into leadership scale and where AI is actually changing how products get built. Welcome to the A list. So, Brian, we've known each other since sfu. Beedie, I'd love to hear your side of the story of how we first met.

We first met volunteering. We coached together. We coached the. The JC west entrepreneurship team.

Actually that was the first team I wanted to coach, except that that team already had like three coaches and marketing had zero coaches. So I just kind of took the, took the team that needed the most support. And then over time I just got more involved in entrepreneurship.

It's sort of a passion of mine. So I think the sponsor opened up or we started guest judging together. And then, and then, yeah, we started hosting practice at your office.

You started coming over to our office. So volunteering was that. And then volunteering turned to settlers of catan. Settlers. Catan turned to brewery hopping.

And that's just sort of how the rest is sort of history.

Yeah, it's been a fun journey, but I think our love and passion for helping others become the best version of themselves, that was something we shared and we were able to see the growth of folks in our community. And you certainly took that much further than I did in the SFU JDC west community. So thank you for that service.

And I think your story of not getting what you wanted originally, but getting it over time speaks to your certain leadership style and your willingness to care for those that you spend time with. So thank you.

Of course, now we've talked about some.

Of that post university connection, how we got. We spent more time together. But tell me about your time at SFU and sort of how that shaped you as a leader today.

Oh, wow. That's a long time ago. So I mean, I, I started volunteering in high school. That's sort of where I got my, my start. Just.

And I'm an eager beaver like lots of other people. Just wanted to build their skills, build their network, build their community. I mean, volunteering for me is.

Was a way to make friends and spend time, good quality time with friends. So starting in high school, when I finished high school, went to university and just like most People like academics was important.

But I think for me like finding a way to give back and just get involved with the community. So I did a lot of volunteering with at the time, Enactus, SFU and the Business, the Business Administration Student Society at sfu.

So those are the main two organizations I work with. So those volunteer opportunities, I think like the classroom experience was good, but.

But I think paired that with sort of the organization experience, what it's like to work with other volunteers, learn from students that were above me, and eventually when I became 3rd year, 4th year, 5th year mentor students that were up and coming in their first and second year, I think understanding how to run a good organization, how not to run a good organization, the students, it was a bit of a place to fail though. Failing was not good, not always good, but it happened. And you know, you should.

I think the biggest learnings always came from like ah, you know, one where we messed up, but two, like always asking ourselves like, how can we be better? So we'd run, we'd run an event or we'd run a program and then you know, we would, my friends were passionate about it.

So we'd always get back, we'd always like be hanging out and be like, how could we do this better for next time? And pass that wisdom on to the next people that are going to run that event or that program. So that's one way that it really helped shape that.

And then the other was sort of Org design is sort of another sort of hobby passion of mine. Like how do you structure organizations to work and operate efficiently?

So large organization, like maybe Your club has 10 people, your org structure doesn't need to be as complex or you don't have to have processes in place. So things seem to be pretty casual.

But once you start working in an organization that has 50 or 100 people, you need support structures, you need processes, you need systems in place.

And having a chance to test out systems in those sort of large organization settings sort of applies really well to figuring out how to structure your structure a business. So as an organization grows from 10 people, like a business grows from like 10 people to 50 to 500.

Like you wouldn't structure it the same way and you need to have systems in place.

And so that was one sort of perspective that I always took from being in student organizations, especially large ones like Enactus, like, like Bass, the student society.

Yeah, it makes sense, right? Like the, the benefit of university is it's a place to learn both in the classroom, out of the classroom.

And the, the cost of failure is much lower reputational risk, of course, among your colleagues. But at the end of the day, everyone's going to move on and get into their career paths.

And I'd rather make those low cost mistakes in university so that you don't make those more expensive mistakes when you're actually in your career or running your own business.

Yeah, I'd argue co op and internships are similar opportunity for students to test and learn ideas because whether they're successful or not, that co op term will end and then they'll be on to their next thing.

So you may as well try and win rather than status quo and try to test out some stuff either on your own dime or on an employer's dime who might let you go later.

So yeah,.

Now you talk about systems and organizations. So over the years I've seen you work for smaller companies and larger companies.

What's been your experience with leadership in the sort of the small context and the large context? And how has your perspective of leadership shifted?

Ooh, good question. So in a small context, I think small, high performing teams are sort of my preferred.

So if you find good people to work with because you can be choosy when you're small, I think when you start to be part of a larger organization, someone else did the hiring, someone else did the recruiting, someone else like may have assembled the team that you're working with. Coming from consulting, you just put together who was whoever is available, whoever has capacity to handle that project.

Sometimes that's how decisions are made, not because someone's good fit for that project. And you have to learn to figure out your ways of working, I think is like the number one thing.

Especially I did a course back in university and it talked about sort of the five stages of team formation, forming, storming, norming, performing. And I think a lot of times in large organizations, you guys just kind of skip to performing without going through those other stages.

And sometimes we forget it. And sometimes the stages, you never get there.

And so when you come to the end of a project that didn't go well, you're just like, why do people hate each other? Why are people yelling at each other? Why is this a bad project?

It's like you didn't kind of go through the phases when you kind of break it down and think about it. And then for the project that do go well, you spend time to get to know each other.

You figure out what people like, what their weight, what their style is, what they expect.

And I think that's From a leadership perspective, I think understanding that how to get the most out of your team, and it's tough when everyone's leader, like, you know, knowing when to lean in and knowing when to take a step back, knowing when to raise your voice and knowing when to shut up is another sort of like leadership skill as well, that I think it's important, especially in these larger group settings.

I think it's very interesting because there's so much M and A happening and different companies joining together with different cultures.

And I don't think many actually spend the time to assess how people in their organizations make decisions and then making sure that they align people with the right not only skill, but also communication style before they start working together.

Everyone's trying to show each other up and show why they're valuable as opposed to how do we work collaboratively together and achieve more as a group? So, yeah, it's very interesting and makes a ton of sense.

I think that's probably why you've had some success in your career, is because you spend that time to actually get to know people before you insert yourself or feel like you need to be the loudest person in the room.

Right? Yeah. And what's interesting that I found for the company, I've been with Applied Digital now for five years. When I joined, we're about 200 people.

I think we're close to 700 people. Five acquisitions.

One of the observations I've made that's really interesting about my time at Apply is while everyone comes from different parts of the world.

South America, Europe, America, the us, Canada, I think the one thing that binds us all together is our love of serving the customer and doing what's best for the customer and our clients. I think we talked a little bit about M and A and how sometimes orgs don't have a good fit.

But if you don't have a shared sense of value, it's really hard to actually bring organizations together to do your best work.

Is there an experience you've had where that maybe didn't go so well and you learned from.

Not that it didn't go well, but learning to work with other cultures was definitely a big one. So when I started working with folks in Latin America or folks in Europe, time zones suck. But Latin America is a different working style.

Even though we share the same values of doing our best work, trying to do our best, doing right by the customer, being right by the client, I had to adjust my communication style a little bit. I had to be a little bit more direct, a little Bit more specific about what I wanted.

Whereas I found sort of in North America you can kind of just give someone a bit less instructions or less. You don't have to be as specific. And people kind of run with that, with that task.

I think that's one of the things I've kind of learned working with other cultures, especially sort of near shore, offshore.

Just the general tendency, I don't want to make it seem like it's like a bad thing, but sometimes you have to be really specific with what you want from people and also provide as much context as possible so they don't just do the thing as you said, but only in that way.

For sure. Yeah. Some folks are maybe better at reading between the lines than others.

And you need to be able to assess to what extent the group of people you're working with will meet your expectations versus those are going to require more expectations.

That's, that's, that's it. Expectations.

Now I find a lot of leadership advice, it's kind of sounds the same across the board. What's one leadership lesson that you've picked up that's perhaps contrarian to what people normally hear?

I was thinking about that because you did a lot of great prep in all this. I was actually thinking it was one of the hardest questions that I actually had to take a step back and actually think about.

And I think a piece of sort of advice that is sort of great leaders is they build consensus, they want to get consensus from everyone. But at the end of the day, you need to make a decision as a leader and sometimes that decision is not like, won't be received well.

And I feel like sometimes as a leader, like you have to, you have to be the one to make tough decisions, to drive tough decisions and then justify it as best as possible.

I think I've been, I've been, I've seen, you know, sort of third hand some, some scenarios where decisions, decisions were made in that way, but they were kind of announced or they were kind of like there wasn't any thought about as to how it would be received by the team and it was received poorly.

And so even though I say as a leader you have to be able to be decisive and make decisions, if you're going to make a decision, be thoughtful about how you announce it, roll it out to people, I think is one of the hardest parts of being a leader.

I think we've seen plenty of examples where the delivery of change was offered poorly.

And the thing about how the world operates now is if a decision is Announced poorly and communicated poorly, the market will determine if it was a good decision or not.

And so I think leaders need to be even more careful with how they make decisions and involving the various stakeholders as well as the communication of the decision to those stakeholders that you can't just unanimously make decisions which if you think of business 20, 30 years ago, that was CEO and there was some leadership and they would make decisions and that was well pushed down. Now there's much more, I would say collaboration with their teams when they push things out.

And I think naturally that's going to lead to better decisions because I think a lot of times people would roll things out, see what worked, if it didn't have to roll back, which is costly sometimes.

Yeah, I think sometimes you have to make an unpopular or very negative decision that will negatively impact a lot of people. But you can still do with grace. You can still do it sort of thoughtfully with empathy.

I think is, is where I've seen like bad news can still be received.

Well, yeah, like that mortgage tech company in the States that laid off all their employees over a zoom call.

Well, and his delivery was terrible. So he became a meme and the company just lost. Huge, huge loss.

So not only did they lose a bunch of good people in that process, but they lost a bunch of customers who saw how their leaders treat people. No one wants to be involved in companies that treat their own people like crap. It's like, you know, why do you go to Costco versus Walmart?

It's like, well, I know they pay their people pretty well and they're organizationally strong and I don't feel like the people that work there are super unhappy. So I'm generally happier to be in a Costco than a Walmart.

Yes, yes.

I think people driven businesses or people, businesses and organizations that take care of their people are, you know, are the ones that the organizations that tend to have a higher likelihood of success. Yeah.

Plus those $50 hot dogs.

Love hot dog.

Quick pause here. If you're enjoying this conversation, go check out the other episodes of the A list.

We've got founders, CEOs and leaders from across Canada sharing the real stories behind how they got there. All right, back to it. So I know you've called yourself a scrappy generalist.

Is that sort of a experience based title you've given yourself or is it something you aspire to be?

No, it's a title I give myself. I think I take pride in being able to be flexible and adaptable to different things.

I think maybe in the past having a specialization served you well.

I think in a world where change is the only thing that is certain, being able to respond to that change, being able to be open to change, is a strength.

Totally.

Don't get me wrong, we still need specialists, people that specialize in certain things. But the term scrappy generalist really resonated with me because I tend to be open. I try to have an open mind.

Which in your line of work, when you're helping large organizations adapt to being digital first in a traditionally analog first business, that's probably incredibly important because you can relate to not only to the tech team, but to the stakeholders that are going to influence the decisions that your team's going to make.

Absolutely. You actually call that a really important thing.

I think a lot of times in tech we just build technology and we kind of forget who we're building it for and we launch something and we're like, all right, push the button, it's live, that's it. And we forget about the change management side of things. We forget about how people feel about changes that are coming.

Because a lot of times you're not. Chances are you're not introducing something completely new.

You're probably migrating an old system or an old way of doing things and you might think it's great. But other people are afraid of, unafraid of change and have a certain way of doing things and they're kind of resistant to it. And we always have to.

Going back to the term empathy, empathize with folks and think about what's their day to day, like, where are they coming from and how will their life change after this change, after the implementation of this new technology or whatever.

Yeah. When technology is moving as quickly as it is, but human tendencies and their base level is not changing at that same pace.

How do leaders grapple with that?

I think sometimes you have to slow down to go fast is an answer to that question, which is if you know that people, if you're getting feedback, especially early feedback, that things are changing too quickly, unless you absolutely have to force it because, you know, financial reasons, you need to move off something or whatever, like try to find your way to figure out a proper sort of migration path or a path to get to people where they need to be.

So I think a term I read was like something like begin with the end in mind kind of Know what good looks like or what great is supposed to look like, and then roll backwards from there and figure out sort of what the intermediate steps are to get to where you expect you want. Where you want people to be.

Yeah, I don't think a lot of leaders do that. They look at shiny objects and problem they want to address, and they're like, okay, let's address this problem. Here's the quickest way to do it. Go.

So, yeah, I think more leaders and teams need to be thinking through the impact of the decisions and the rollout.

I think it goes back to that question about unpopular leadership decisions, because sometimes you have to make an unpopular decision. Going back to the whole making a decision, even though people are resistant to it.

I think it's okay to make those decisions, but just make sure people feel heard and say that this is a decision and I own it and this is where we have to be. Yep.

Yep. Delivery.

Yeah.

Now, you did your master's in education and you've been involved in some edtech platforms in the past. How do you think becoming an educator is important in technology today?

I think it goes back to sort of the change management piece. And from an education perspective, the term I like to use in the business context is enablement.

How do you enable your team, your employees to get to where they need to be?

And education plays a part of that because I think some of the educational principles I learned in my masters are around the term they like to use in education. Scaffolding, I like to use the term context is another thing.

So how do you give people enough background, enough explanation as to what they need to know before you explain the thing that you want to teach them? And then you explain the concept, and then you have to give them an opportunity to practice that.

So there's different models in education space to get people to. From not knowing what it is they were before and learning that new concept.

Got it. So basically, it's more.

Whenever there's change happening, you want to make sure you're educating your team or your stakeholders on how to get from A to B.

Yes.

So that they can actually use what you've offered them or built for them.

Yes.

Yeah.

Yes.

It makes sense. Yeah.

Yes. And it's interesting. Like, a lot of training is like, oh, I gave you a workshop or I gave you a demo, and then that's it. And it's like, good luck.

It's generally not enough. It's not enough.

You need to give people time to practice and give people time to feel comfortable, ask questions and oftentimes I'm like when I give demos and trainings all the time and questions come like weeks or months later. So anticipating that is also just.

It's hard in consulting because contracts are fixed length but if you're on a retainer, you have a long term relationship with a client. Just being able to support them with those questions are important.

Yeah, I think my software product advisor Flow, we do regular practice management events where we'll educate on various topics, but then tie it back to software so people can actually understand use cases and where functions and features fit into workflows.

Because you're right, if you just give someone here's a tool and they can't think of their regular daily work and how this fits in, then you've just sort of given them something else on their pile of things to do which isn't helping them be more efficient.

The general. What I've heard and actually, I've actually, I said this earlier today actually in a meeting.

I'm like, if you're going to ask someone to adopt something, it needs to be at least 10 times better than how it's already done. Otherwise it's going to be a lift, an effort to get to the adoption of the usage of the technology or the thing you're trying to roll out.

100%. Yeah.

Because that time to change, people see those barriers, they're not going to bother to learn to become better if they can just keep doing what they're always doing. Inertia is just so powerful.

Yes, exactly. Oftentimes in technology we think, oh, different technology platforms compete with each other.

Actually the competition is sometimes not doing anything at all.

For sure. Yeah. And. And that works. There's billion dollar companies making their money off inertia.

Yeah, exactly.

Now you're someone who likes to champion change for the good. For good. You were president of your Strata Council and led a major redevelopment.

I know there was some, a lot of ups and downs with that project, but what did that experience teach you? About what? Patience, persistence and community.

Ooh. That's also digging deep in the archives. It's been a while since I've been in that role.

I think from what I remember, you know, I originally joined my. I think everyone one. I think if you're, if you're young and you're part of your, and you own your condo, your own space, you know, if you can.

I advise everyone to join their Strata Council, even if it's not in like a, like a high leadership role, just as a member, just to see how Things are going because it's your investment and you should, you know, be part of that and have at least a say in it for sure, if you have the time.

And you won't get more time when you become a parent, so. Exactly.

Exactly. I did this when I was young and single.

I showed up to a Strata Council meeting one day, maybe an AGM or something, and I looked at the Strata Council and people were yelling at them and our AGMs were like a gong show. They would go on for like two hours and people are yelling at the council and our property manager and people were exhausted.

Look on everyone's face at the council was exhausted. And I just kind of put my hand up and said, I can help. And then all of a sudden I found myself on council.

That's just kind of how I fell into it as part of, in terms of, like, of how I got involved. But I think selfishly, I got involved because of that.

Like, what would happen was they'd say, oh, here's an update with that's happened in the building, and then we'll give an update in three months. And then six months would go by and there would be no, no communication about this. And so that's no way to build a community.

It's no way to build trust.

No.

And so one of the takeaways I took from, from running my Strata Council was just over communicating. Over communicating. Being super transparent, being super simple about how you're communicating.

So sometimes I get a document, it's like really, like two pages and super dense, and it's like, no bullets. Like, no bullets, no bold, no titles. I'm like, this is an essay. I can't. I can't read this.

So make sure your communication is clear and you're outlining sort of all the different specific things. I think I didn't do everything. This is public. I didn't do everything, like, according to, like, how you should be running a council.

But I often found, like, getting everyone's email, emailing everyone frequently, especially if things are changing. Sometimes I'd email the ownership base like twice a day, three times a day if something had changed. Just keep everyone afloat.

Printed signs all the time to leave in the elevator. We only had one elevator, so everyone had to take it. So signs in front of the elevator, signs in the elevator, hard to miss.

And that over communication and giving people a path, there's uncertainty. Like, I don't know how things are going to go, but it can either go this way, like path one, path two, path three.

And here are the pros and cons of each. And here's how we would make that decision depending on different things that happen around in our building.

And so that's from what I learned from that was to over communicate, to build trust.

And it got to the point where after doing that for maybe a couple of years, you know that you hear the stories about people that have a bunch of proxies and they always vote things down at some point. I think of our like 60 units I had like 40 proxies that people, if I just emailed people I would get all the proxies and people just.

But also it was Covid during that time so people couldn't get together. So a lot of people just pre emailed their proxy to me and they would agree with any sort of high level decisions that needed to be made.

President Bryan in action.

But, but what I love about this Brian is you, you took a problem council and you turned it into a functioning council and you actually got some major decisions pushed across that benefited the whole building.

If you think of what the real estate market is now, if you didn't get that deal done, people would be holding onto something that was worth far less than you got them.

Oh yeah. For context, our building had millions of dollars in repairs because we were part of a leaky condo that just didn't get fixed.

And so I bought in knowing this. So I'm like, ah, it's a future Brian problem. Well, future Brian was past Brian like five years ago.

And while you're thinking about questions, this concept of over communicating, I think one time we just had a product launch actually this is really fresh.

We just had a product launch and we over communicated and someone was like, I didn't need to be part of this communication for when this website went live, it was like a VP or something and I was like, I don't know what I would have changed because that person could have left the group chat at any time if they didn't want to get any messages about all the things that were happening. I actually think maybe in the future we shouldn't invite that person. We have to learn because we didn't know at the time.

But we over communicated everything that was going on because it was such a mission critical site. Applying what happened in strata, applying what happened for a product launch or technology launch, website launch, same sort of principles, whole.

Well you're dealing with stakeholders, right?

Yes.

And you've had a good idea of say the demographics of people in the building and so you thought, okay, they might not Read their email. So we need paper to be visible for those people. You still want to do the emails.

So the people that perhaps choose not to read those things, you hit them.

So you're not thinking through the channels, thinking through the demographics of who is going to be impacted by the change and then getting them involved and participating in the decision making process. And that's, I think that sounds like a good leader to me. So well done.

Now I know you've talked about balancing what's best for customers with what's best for your team. How do you make those decisions during perhaps tight fiscal times like what we are in right now?

It goes back to setting expectations.

So oftentimes like where this sort of where people get really stressed out is like delivery timelines and delivering near impossible or impossible dates. I recently had this on a client. They wanted to go live with that new site that just went live. They wanted to go live in June.

There were a number of delays, most not our fault. Actually most of them were not our fault. But I could anticipate that based on the things we needed.

Building technology long enough, you recognize the patterns when you have dependencies downstream. Like, I cannot rely on things I cannot control. So I have to plan buffer for that.

And as a result, like I kept telling my project manager, I'm like, we could do it if everything is perfect. If everything's perfect and everything works perfectly, we can meet our whatever June date that we set.

But I'm like, if you want to give ourselves buffer and set expectations right, we should probably push this to July.

And that's an example of me standing up for the team, making sure the team like otherwise the alternative was, oh, the team might have to work overtime, which they were already doing. The team might have to work weekends, which they would have had to do if we didn't push, push the timelines.

And even then, like sometimes for when you have dependencies, you can't work weekends. Like you're just sitting kind of waiting for someone else to do their part so you can do your part totally.

So it's throwing more hours at a project is not going to make it go any faster.

Sometimes it's just the one of the wisdom of sort of building technology is anywhere there's an integration or anywhere you have downstream dependencies on other partners or services, don't expect them to be 100% functioning when it reaches your hand, 100%. And so you have to plan for that. And you know what? You set a date. We set a date in July.

We hit it but you know, in theory we could have gone sooner, we could have gone live sooner if everything fell into place, which it didn't.

So I remember like hitting the, Remember going live on the date that we said we were going to go live and messaging my project manager and saying told you so. Told you so.

Yeah, well we've seen that within my software product as well where we'll set a timeline where the devs actually set it in advance and then we go and get closer to that date and it's not being done.

Sometimes we do have to work over time, but if it's a self imposed date where they projected out how long it's going to take them and there was no external dependencies, then I think there's some understanding as to why they may need to work overtime because they can't meet the deadline that we have promised to other people outside of our organization. Yes.

And that's different. Working for an agency versus client promises because it's egg on your face, it's egg on the company. It's a reputational risk.

Correct.

And so if you set a timeline and you kind of.

Software developer is hard, software development is hard because you're doing things that have never been done before oftentimes and maybe you didn't get all the right requirements or you ran into an issue that you didn't realize. But yeah, if you have control over things, then the other thing I like to preach from a leadership perspective is holding people accountable.

So that's a good reminder. Like the way I like to run standup. I don't know if you do standups, but often standups are like, oh, what'd you do yesterday?

What are you doing today? What are you blocked on? And some people just go and give their update and then it's like nothing.

And then I like to bring up the board and say, okay, rather than going around the table, like to bring to the board each individual task. All right, Brandon, this was assigned to you. I leave comments on my tasks.

So if you said, hey, I was going to get, it's Tuesday and you know I'm planning to get this done for Thursday. Wednesday, Yeah, I said I was on track. Thursday comes. Nope, I'm not done. What changed and how come you didn't raise it before?

And it's not to shame people, it's to keep people accountable a hundred percent. So that's another sort of style. I like to run standups.

Oh yeah? Yeah. I don't run our dev standups, our tech lead does. But Jira board assigning time to tickets. It's not to, you know, keep our.

We're not trying to micromanage people, but we want them to be accountable to the work they're doing as we need to be accountable to our customers. And ultimately the times that we develop, if we do it effectively or consistent amount of time, they're going to talk to other advisors about that.

So it percolates. If you can deliver on promises internally, externally, that will be visible and people will notice and support you more.

But going back to balancing the business need versus the team need. So in that example where I gave, where it's like it's Tuesday, gave an update Wednesday, Thursday hits fan, it's not done.

If there's a reasonable explanation, I'm happy to go back to course our stakeholders happy to go to the customer and say, hey, we hit a hiccup.

I can manage expectations early, but if you kind of get it too late and that Thursday all of a hit, all of a sudden hits next Monday and then all of a sudden your timelines have slow slipped for sure, then it's not. That's not a good look either.

100%. Yeah.

Yeah.

The transparency is key across the board.

Yes.

So AI is sort of eating up the airwaves across the board in tech.

Where do you see AI changing our businesses and how is it going to change leadership styles, considering how transformative, quote unquote, it is and how embedded it's becoming in all of our regular workflows?

Ooh, okay, Business and leadership. I'll start with business first. I started using a very.

So for those that may be using AI, I think the number one thing to think about is what model are you using?

Because the model that you're using will be either will be the difference between I prompt you once or twice and I get the output I want, or I have to prompt you 4, 8, 12 times to get the output that I want. Lower models, you have to prompt more. Higher models generally at the prompt less. But that was my general experience.

I think I started using Claude in February and the outputs were so good. They're actually really good. And so that was the moment in February where I became a believer in AI.

Not that it would take everything away, but that it was competent enough to do a lot of the tasks that are sort of tasks that we don't really want to do. Artifact creation, synthesis, creating presentations.

I think those are sort of the tasks that there's a lot of things that I did that you would normally had to do it manually, would have taken Days and I did them in minutes, like set the prompt, spent some time to actually figure out what I want, hit enter, walked away, came back. Now you have to spend time to review it.

I don't think we're at a point where you can just let your agents kind of do what they want, but you do have to review every output.

I think, especially when it comes to using agents in a collaboration setting, I can't just take my output from my agent and then just email it to you or slack it to you and say, hey, here it is. Otherwise it's putting undue stress on you to have to like read it and go, this doesn't like, well, how come this doesn't make sense?

So there still needs to be a human sort of reviewer. And so we're moving away from creating to actually reviewing and judging and applying taste.

I think that's sort of where AI is going from a business perspective. I think it's just going to be a tool like anything else. Like rollback. Before email was like a thing like how do people communicate?

Phone calls, letters. Like all of a sudden we have email. Now everyone has to use email. All of a sudden we have teams and everyone has to use Microsoft Teams.

Like AI is just going to be another one of those tools where you'll be expected to use it and everyone will just have to figure out how to use it for their everyday tasks and figure out what it's good for and figure out what it's not good for. And that'll only come from experimenting. And it's tough in this. It's tough now because you might run an experiment. Today is July. August.

We're already in August. Holy smokes. Today's August. You run an experiment in August sometime.

If you run the same experiment in October, it might drastically change because the models are consistently changing and updating. And so you can't just test it once and then be like, nah, it's not good and walk away and never do it again.

You have to be constantly coming back and trying.

And I'm not saying like throw like hard everyday tasks at it, but like, you know, if you're going to spend a couple of hours doing something, maybe just see if the agent can do it first and spend some time doing that first. That's my default now. So if I have to build a presentation that's like, I know it's a 10 page slide deck or whatever.

Actually this is exactly what happened last month. I had to build a 10 page slide deck or whatever and I was like, I Wonder if AI is good enough to do this.

And I ran it through and it was like, here you go. And I was like, oh, that's pretty good. It made a whole PowerPoint that's, you know, 60%, 70% of the way there.

And interestingly enough, one of the things I have to do now is decide how much do I keep reprompting versus how much do I. Am I good enough? Is output good enough? Where I just have to tweak things?

Right.

I think that's where I'm at now. And it's a cycle of, I guess, what's the term? Not roll back but like you have to circle back. So what will happen is I'll create an Artifact.

It'll be 80%, I'll do the final 20%, I'll send it to you. You're like, I have some more feedback. And you give more feedback because we're collaborating. It's just my agent. You can't prompt my agent yet.

Then I take that feedback and go, okay, do I do it now? I have to decide do I do it myself or do I re upload the changes and reprompt? And that's sort of the cycle of at least day to day work.

I'm not talking about like automated workflows and things like that.

I'm just talking about like artifact creation, which I think a majority of people are using to like create documents or synthesize like large amounts of information.

I hope it from a leadership context because there's a lot of layoffs happening, a lot of people worried about their jobs. How do you see AI changing the job market, especially for younger people that I think are probably a bit worried.

It's interesting that you.

I just, I was listening to a podcast yesterday and I believe as I was like, I was listening to like the chief like product technology officer at Netflix, and Netflix, I don't think used to hire interns and they're like, oh yeah, we started hiring interns again. And I was like, I wonder why? And they're like, well, because one, they're malleable, so they're kind of open minded.

You can kind of mold them to whatever you like. But.

But two, they also understand culture in a way that a 40 year old or a 50 year old who would consider senior, if you're only going to hire senior people, they come with that perspective already pre baked and they might not be aware of trends or might not be as locked in. And so I think it goes back to the principles, first principles of having diversity. You should try to have as much diversity as possible.

So you have diversity in thought.

So you need pressure test and check all the things that you're doing for, especially if you're serving sort of a wide audience like Netflix, but if your demographic is like, you know, a smaller or narrow band, then maybe you don't need that perspective. But I think it goes first principles, it goes down to making sure that you're having diverse thought.

And then I think we also forget that young people are digital natives and, you know, young people probably know how to use AI better than I do in some cases, and so we can't take that away.

So, yes, I'm worried about young people and not being able to kind of get the experiences that we had kind of coming up, but it's just going to be different.

Yeah, no, I'd agree with that. I've gotten great value from the internship programs that we have and as many have grown into for permanent roles within our company.

And I think we've become a leader in Canada as it relates to our digital engagement and how we leverage digital tools to engage with clients and prospective clients on the Internet. So, yeah, I'd agree with you, and I think you're right.

There's.

There's great opportunities for young people, even if some of the large companies aren't providing these internship programs.

That's good news to hear companies like Netflix are bringing them back because I do think it's, it's an incredible opportunity for growth and then to build your team from within as opposed to feel like you needed to poach people from elsewhere.

Yes. Going back to first principles. So I think a lot of people are like, oh, we'll just use AI to solve our problems. It's like, okay, well, rollback.

What problem are you trying to solve? And have you taken the time to map out your daily, weekly, monthly tasks and things you need to do?

And sometimes we don't take the time to really go, why do we do this? Why are we doing it?

So when it comes to things like technology and software development, you kind of go through discovery requirements, gathering, design, prototype development, user testing, test, qa, deploying. Why did we follow that? Going back and thinking, why do we follow that cycle in the first place?

And if we're going to replace it with something, how does that change? Are we replacing one part of that process? Or do we have to reimagine that process from the ground up?

And if we imagine that process from the ground up, what does that look like?

And we have hypotheses about how things might work, how do we experiment or how do we trial that in a way that will give us trust and confidence that if we implement this new process using AI, that it will be as good, if not better than the old way of doing things.

An example of this might be like, I can envision a world where like before, a lot of times when you're building websites, people are like, ah, the designs aren't ready, the designs aren't ready.

And all of a sudden now the ball in the neck was not to put any constraints on designers, but like the bottleneck was like, how fast can someone put together a design that was approved? Not only is the design has to be approved, but like if like I made a design then you, the boss has to approve it. That still takes time.

That's not something an agent can quickly do.

But now with AI, like rather than presenting you with one design, you could present 10 designs and you could use like synthesized testing to have AI pretend to be certain Personas about who you're building this product for and then run the test through, run all your prototypes through that synthetic user testing. And then I could come to you, the boss who's going to make the decision on the, on the design and say, these are the 10 designs.

We kind of, we explore 10 designs and here are the three that, that did the best and here are the pros and cons of each one. And then a lot of times like decisions are made by gut, which is true, a lot of decisions are always still being made by emotions.

But we can now use AI to try to make a data informed decision.

I like that.

And I think from a design and product development perspective that makes a ton of sense in terms of how the innovation is going to impact the process and make things run more smoothly.

But if we look at perhaps more regulated industries like say Pharmacare or Pharma or Fintech, how do you see these innovations impacting those industries where there might be more privacy and steps involved in order to integrate the technology?

It goes back to how you prompt your agents. And so if you operate in those spaces, then you have to give that context to your agent.

It's the same as like any, like if you're going to hire someone in pharmacy, you would tell them pharma or in Fintech, you have to make them aware that these are, these are the different compliance and rules. It's the same with AI.

So if you, if you, if you give them the context and you provide that, that background and say we're operating in this highly regulated industry. Please keep this in mind. What rules or laws or jurisdictions do we have to be mindful of?

And they will tell you, and then you say, great, can you please make sure that everything we do adheres to these principles? And then you, everything that, all the prompts that you run through, it should have those lenses in mind when you are trying to do any of your work.

Can we trust the agents are going to follow the rules all the time?

No, not yet, I don't think. I mean, maybe some people have set those rules or have set those contexts and they've gotten to a point now where you can trust.

But the same goes for working with new people. Like, you'll give me an assignment if I work for you and I'll do it and you'll be like, I don't like it this way.

And you'll give me feedback and then so like, okay, the next time I'll do that same task, come back, I'll have that feedback in mind. Give it to you. You forgot about that compliance that we talked about. Oh, okay, run it, run it, do it again, Give it to you. And now.

And if the output is good and I do that consistently, then you'll trust me and you won't have to reprompt me or give me guidance. I think the same goes for AI. Like we think, oh, providing guidance is a one time thing. And it's not.

It's like you need to train your agent to eventually get to the point where you can trust it to maybe eventually operate autonomously. I think that's sort of the direction where that, where that's kind of going.

But you have to have, you have to see it in action enough times to give it the trust.

Yeah, and I'd agree with that.

With a lot of tasks, I think the big challenge, particularly in an industry like mine, is where we are entrusted with a ton of confidential private client information.

And so the risk of a breach or of information somehow ending up in a search engine that was entered into an AI tool that was supposed to keep that information confidential simply because someone checked a box that said don't train the model. That's where I think big risk is. That's perhaps underpriced for many that are not educated on the subject.

I agree.

Actually, on the way here I was listening to someone who had, I think offline AI models are eventually going to take over, which is fascinating because we started on prem. All this stuff had to be in a server that lived in your server room and then everything migrated to the cloud, everything's there.

We trusted the cloud to keep all our stuff secure and private. And then our agents are now in the cloud, but we're like, I don't know if they'll actually be private.

Right.

And then now we're actually reverting kind of back to an offline model where you're just like, hey, this should never leak. Like, it'll only do what I want. I don't, I don't have to be at the mercy of a tech company that is giving certain guardrails or certain things.

And so it was just thought provoking that at some point.

And maybe I think Apple will sort of take this sort of approach where your model will be offline because you want it to be your stuff and you want to keep your stuff private and contained on your own device.

And then at a macro scale, you look at countries like Europe and Canada that are really trying to build up their data centers so they're not relying on the US for, for their AI models and services.

Yeah, Apple's probably the best to watch because they've done a good job of keeping a closed ecosystem, which has been very profitable for them, versus, say, our Google or Microsoft counterparts, which are more sort of open.

Different. Different business models.

Oh, definitely.

Different business models that cater to different people. Some people care about their data a lot, some people don't. Yeah, well, they care about their convenience.

More than they care about their.

That's right, I see. Yes, exactly.

Brian's been really interesting. We've talked about the education, arc, product management bit of your view on AI.

If there was one or two things you'd want people to take away from this conversation, what would you say they are?

Okay, I'll start with AI. I think for one point around AI is just to keep testing and keep pushing and keep experimenting.

I think it goes back to the first principles of always being curious. It goes back to my whole being a scrappy generalist.

It kind of all sort of aligns when I think about it, just always be trying new things and pushing the boundaries.

I think that's the first and I guess the second is, I guess when it comes to leadership, leadership style, just leading with empathy and just taking some time to think about.

If you're deploying tech or you're serving customers or you're changing things for other people, your customers or your users, what's it like to be in their shoes and just take a little bit of time to think about that and empathize with that before making changes or have that be in the back of your mind as you're rolling out that new technology,.

Both of which I think are incredible lessons for those on the call and also for me.

So I appreciate you for leading with empathy in our communities, being a champion, and also challenging all of us to continue to stay curious and not get stuck in our ways so that we can inspire the next generation to hopefully surpass us, which they will. Well, if people want to find you, Brian, where's the best place for them to get in touch?

Oh, I have a website, but it's been under maintenance for a long time. Maybe I should get AI to revamp my website for me. You can find me on LinkedIn.

Okay.

Yeah.

Brian Yanmuk, thank you so much for sharing your time with us today.

Thanks so much for having me.