The Digital Project Manager

What Non-Technical PMs Really Need to Know About AI

• Galen Low

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 40:06

AI is touching every role, every industry, and every level of an organization—but that doesn’t mean every project manager needs to become an engineer. In this conversation, Galen Low sits down with AI strategist and engineering leader Kesha Williams to talk about what PMs really need to know to stay relevant in an AI‑infused world.

They explore how to lead AI‑related projects without getting lost in the technical weeds, how to confidently translate between business and technical teams, and how to focus on outcomes over hype. With clear examples, sharp insights, and a dash of humor, this episode is packed with guidance for delivery leaders navigating AI today—and preparing for what’s next.

Resources from this episode:

Galen Low:

What are your top tips for project managers who need to act as that translator between cross-functional teams that are working on or using AI?

Kesha Williams:

Instead of chasing tools, really think about the business outcome. What are the problems we're trying to solve? Focus on the what first, then you focus on the how.

Galen Low:

What is your take on certifications around AI and why someone might get them, and would it be worth it?

Kesha Williams:

I love certifications. Certifications could get you the interview, but they will not get you the job. It's important to have that hands-on piece and online portfolio where you showcase the projects that you've actually developed using those technologies.

Galen Low:

Does every project leader need to become a systems engineer in order to stay relevant in an AI boosted economy?

Kesha Williams:

That is a great place to start, and I would say no. For me, it's more about—

Galen Low:

Welcome to The Digital Project Manager Podcast — the show that helps delivery leaders work smarter, deliver smoother, and lead their teams with confidence in the age of AI. I'm Galen, and every week we dive into real world strategies, emerging trends, proven frameworks, and the occasional war story from the project front lines. Whether you're steering massive transformation projects, wrangling AI workflows, or just trying to keep the chaos under control, you're in the right place. Let's get into it. Today we're talking about how project leaders can become more effective translators when facilitating conversations about AI between tech and business stakeholders, and also how that might be the key to de-risking that looming prospect of being just another AI rollout that failed. With me today is Kesha Williams, Founder and Managing Partner at KeySoft, a consultancy that helps organizations adopt AI through strategy, engineering enablement, and developer engagement. Kesha herself has a deep history working at the intersection of AI strategy, engineering, education, and influence. She has over 25 years of experience across software engineering, enterprise architecture, and AI innovation, which she uses to deploy AI in complex real world environments. She has also recently turned her training programs to less technical audiences and has finished her first 15 day AI bootcamp in partnership with LinkedIn Learning. Kesha, thanks so much for being with me here today.

Kesha Williams:

Thank you, Galen. I am super excited to be here and have this conversation.

Galen Low:

I'm excited too. I love what you do. You put so much energy into your work. You are a busy body. I don't know how you do it. There's so much education, just like streaming through your LinkedIn feed. I know you've got a number of different LinkedIn learning courses and other courses as well. I'm really excited to dive in. I know your audience is mostly, it's normally more technical audiences and you've kind of been branching out into sort of the le us le folks.

Kesha Williams:

Right.

Galen Low:

I really appreciate that. I do hope that we kind of zig and zag today, but just in case, here's the roadmap that I've sketched out for us. To start off, I just wanted to like set the stage by getting your hot take on one big juicy question that my listeners will probably want to know the answer to, but then I'd like to unpack that and just talk about three things. Firstly, I wanted to talk about why it's important for non-technical PMs to understand AI. Then I'd like to explore ways that PMs can be good translators between technical and business stakeholders when it comes to AI, and also ways that they can improve at that. Lastly, I'd just like to get your perspective just on the future of the project manager role when it comes to projects that involve AI or are producing an AI native solution.

Kesha Williams:

Sounds like it's going to be a great conversation.

Galen Low:

Ambitious, but we're gonna do it.

Kesha Williams:

Right.

Galen Low:

All right. I thought I'd start us off with one big hairy question and just to give it some context. A lot of folks in my community are feeling, I don't know, more than a bit anxious I'd say about AI and how it changes expectations around their current role and their career trajectory and the skills that they need to focus on. So my big question is this, does every project leader need to become a systems engineer in order to stay relevant in an AI boosted economy?

Kesha Williams:

That is a great place to start, and I would say no. And if you look at what's going on in the industry now with a lot of these AI, no-code and low-code tools, you do not have to even know how to code or have a technical background to build an AI agent or to even build an app. For me, it's more about understanding like the systems thinking, what is AI? What can it do? How can I apply it in my current role, how can I apply it on a project? So no, you do not have to become a systems engineer or even become very technical. In order to like stay relevant and understand how to use AI and apply it.

Galen Low:

Boom. I love that and I'm sure everyone listening just sigh a sigh of relief. But also I really like that inflection because like, I don't know, I find that like this no code, vibe, code kind of ecosystem. Or at least culturally there's this pressure to be like, you don't need to code, have outer. But I like what you're saying about it still takes a systems thinking mindset to approach this. And we see it all the time, or at least I do in my circles, folks who are like vibe coding and they're like, wait, I need to debug and QA still. I'm not writing code, but like it is the sort of like that IT engineering piece, right? Yeah. It's like, oh wait, I still have to know what I want this thing to do and how, right, like it's.

Kesha Williams:

Right. And once I put it in production, you mean I have to update it, I have to enhance it, I have to maintain it. I thought I was done.

Galen Low:

How about vibe DevOps? Honestly, I think that's like, you know, it is a great place to like kick off from because I think. Then the question is how much is the right amount of technical or how do you kind of acquire that sort of systems thinking? And I wanted to get your perspective on this because you yourself are very technically accomplished. You're an engineer, you're an enterprise and solutions architect. You've led teams of developers and you've built an incredibly large and engaged audience around exactly that. Recently, as I mentioned, you've been expanding your audience to us less technical folks. For example, that 15 day bootcamp that you had in partnership with LinkedIn Learning, it kind of led folks through a bit of a journey into AI. Actually quite deep, starting from like pretty far back into some pretty deep stuff. Why is it important for you to bring your education chops and your AI knowhow to the masses?

Kesha Williams:

It's important for me because AI is touching every industry, every role, every level within an organization, and I don't want anyone to be left behind. When I look at the power of AI and what it can do around boosting productivity, automating workflows, it's just a very exciting time to be in tech, and I just want everyone to embrace AI and figure out where it fits in their day to day.

Galen Low:

Are you finding yourself encountering some like false assumptions and myths and like misperceptions around AI? Because part of what you do is kind of being like, okay, some of these folks who might not already have foundational knowledge in terms of assistance thinking or even have done, you know, had time to do all the research on what AI can do. Are you finding that part of it is kind of cutting through some of the mythology or some of the, like fear-mongering around what AI is and how it like applies to folks' job, whether or not they're technical?

Kesha Williams:

Definitely, I would say like Hollywood and all the science fiction movies have not helped. It's made my job a whole lot harder, but my goal, and it's really always been a passion of mine to demystify AI and machine learning and just making these complex technologies just very approachable. But like I said, Hollywood has not helped at all.

Galen Low:

It's like the Hollywood thing and then there's like this slew of articles, and don't get me wrong, like credible, but also framed in a way to get eyeballs. You know, where it's like former head of AI at Google says, AGI will take over in seven years. Or you know, all these people who like are the AI authorities. It's sort of framed in such a way where it's like, oh my gosh. But also in a way it's like, of course these individuals have thought about all of this. Part of what they are is the forefathers and foremothers of AI have had to think about what the implications are. Because when you paint it out on a far enough timeline, it is going to change things a lot. And it goes all sorts of different ways. But even just like what you said is really relatable, it's like we all need to get kind of on the same page, at least maybe not all at the same level. It is impacting everybody and their job and expectations around their job. And the better we all understand it or the better we have like a shared vocabulary for it, we can actually like talk about it. It's not the same as when I first started. My background is in digital project management. The development teams that I work with, they were like front end and backend engineers and I remember the first few days, right? I'm like, none of that makes sense to me. You guys like, yeah, go deploy. Do that thing with git. I'll see you later. And that learning curve of, you know, just being able to relate and actually have a meaningful dialogue instead of just looking at something and be like, yeah, okay. Do that regression testing or whatever it is. And and then I'll see you later. How have you been finding some of the training that you've been doing with folks? I think you have your everyday AI chorus, you just have that 15 day bootcamp. What are some of the like impacts or insights that you've seen from your learners?

Kesha Williams:

Right, and so like you mentioned, I kicked off the everyday AI 15 day AI challenge in partnership with LinkedIn Learning. Today is actually day 15, the last day. Yeah, so I'm super excited about how everything turned out. I've gotten a lot of feedback from learners where I've seen apprehension around AI turn to excitement. I've seen where people now realize where they can fit AI into their day-to-day and use it to actually save time and boost productivity. One learner reached out and said that she used AI for a task, and it took her maybe 30 minutes to do it, and otherwise it would've taken her several days to put the outline together. And so things like that, it's. The reason why I created the 15 day AI challenge, and like you mentioned, it's really for non-technical people to really start to understand just basic concepts and foundations around AI and then how it can be applied in their everyday life.

Galen Low:

I love that. The sense I get from it is that it's quite hands on. It's like, oh yeah they're actually doing it. It's funny. So bit of a segue. Last night we were playing a card game, exploding kittens, and the instruction manual says, don't read these instructions. This is actually the worst way to learn how to play a game. Watch this video and then play around. And I'm like, actually, you know what? That applies to AI so much. I think a lot of the folks who have, and myself included, I had the most anxiety around AI when it was just pressure and I hadn't opened any of the tools yet. And it's almost like that inbox, right? You look at the number on the red badge and you're like, oh my gosh, it says 967. That's crazy. I don't even wanna look at it. I'm intimidated by the number. Right? I haven't even looked at what is in my inbox yet. It was that moment for me. But I love that idea of just kind of like. Guiding people down that path through the foundations, and it gets pretty deep. Like what are folks doing today on day 15?

Kesha Williams:

Day 15 is all about creating your first AI agent without code, you do not have to know how to code. It's all through a drag and drop interface to build your first AI agent. So everything worked up to day 15.

Galen Low:

I love that. It's like day one, i'm making this up, but day one is like, what is AI? Day 15 is build an agent.

Kesha Williams:

Basically. Basically, that's how it went.

Galen Low:

I really like that. I really like that. I think that's maybe a good segue into like the meaty topic that I wanted to dive into, which is how less technical project managers can de-risk their AI boosted projects. By having the right conversations, the right conversation to drive alignment around AI with both technical and business stakeholders, and I guess everything in between. I know like we're so used to polarizing it, it's like technical or not technical, but I guess what I mean is translating across different stakeholders who know different things and have different POVs and have different skill sets. First of all, should it even be the project manager's role to play translator for conversations around AI? Like maybe like validate that first. Is that a good idea, bad idea in your mind?

Kesha Williams:

Oh, it's definitely a good idea. Like project managers are already doing translation, like translating project risk, timelines, having those conversations. So I think it's just a natural fit for PMs to serve in that translation role when it comes to AI.

Galen Low:

It's a really good point. Yeah. There is a lot of translation already in the role.

Kesha Williams:

Right.

Galen Low:

And actually, you know, it's funny 'cause you and I were having a conversation leading up to this. You had a post about agents and MCP being the new APIs, and I'm I'm going back to my sort of early baby project manager days where I'm like, what's an API? But it was that thing that I, you know, as a means of integrating systems, plugging things together.

Kesha Williams:

Right.

Galen Low:

It was very much that thing I needed to be the translator for to explain to some of the business stakeholders, some of the folks who. Didn't necessarily understand APIs, to kind of educate them on why it's important that we take these steps, you know, why we build our data dictionary, why we explore the documentation, why we have to do a little bit of a an MVP or like a proof of concept. I see it kind of being the same and yet also different because. There's so much coming down the pipe around AI and there's so much changing. I feel like there's a, like a new model introduced every day, right? There is around, you know, the l lm, the classic LLMs or the image generation and all that. What are your top tips for project managers who need to act as that translator between cross-functional teams that are working on or using AI? And can we maybe even go through a few examples of like. What an impactful conversation might be that could either go sideways or could also make a project go way better.

Kesha Williams:

Right. So my first tip would be not to chase the tools like you mentioned. Like literally every day there is going to be a change to an existing tool or a new tool that comes out. So instead of chasing tools, really think about the business outcome. What are the problems we're trying to solve? It's really that principle that has just been around in the project space. Focus on the what first, then you focus on the how. And so like, that's my top tip, really think about the outcomes and don't chase tools because you will go down a rabbit hole and you'll just go all off track. That's the top tip that I have. And then PMs don't really need to be technical. But understand the AI concepts. When someone says a model, you understand what that is. When someone says prompt engineering, you understand what that is. When someone says agent, you understand what that is. So just really understanding these high level concepts and how they fit into the how of what's being developed. So that's the first part of your question. And then the second part of your question was around like a conversation.

Galen Low:

Yeah. What are some of the conversations that just, maybe some examples, doesn't have to be the defacto big AI conversation that's gonna make or break your project, but, you know, along the way we're having conversations with folks and you know, I've seen even pre AI and actually today, you know, seeing conversations that kind of go a bit sideways. For some of the reasons that, you know, we're talking around where like a PM might just not feel confident about like a concept. They might make it up and then they might regret it. You know, like it's those types of things. Or like, you know, trying to like overly simplify. Yeah. I don't know. I thought maybe we can riff, I'll riff too.

Kesha Williams:

Yeah, sure. So let's say you're a PM for a project. Of course it's an AI project. And you have a machine learning engineer on your project and they come to you and say, the model is ready, accuracy looks good. Okay. So I, the way that conversation would go sideways, the PM would be like, okay, let's go to production. You know? I think there needs to be additional questions from the PM related to, okay. First, the PM has to understand what is a model. They have to understand what is accuracy. So accuracy is the way you judge how the model is performing. And so one conversation or one question could be, what does accuracy mean in this context? Another question could be, and I sort of alluded to this earlier, once the model is in production, what is the maintenance and the retraining process? Because specifically with AI models, AI systems, once you put it in production, there's this issue that can. Pop up called Drift model Drift, and that's just where the data in the real world changes over time and it could impact the quality or the output from the machine learning model. So it's important when you put a project or an AI project in production to have that retraining, ongoing maintenance conversation because that will impact just the overall output of what the system is trying to provide. So just asking the right questions when it comes to like these AI systems that we're putting in production.

Galen Low:

I really love that point, especially about Drift because. I find that, you know, even the quite technical project managers that I know and that I work with, we still come from this world where things are pretty deterministic, right? Right? It works this way in testing. It works this way in UAT, when we push it to prod, it's gonna work that way and it's gonna work that way every time until we build something else and it breaks it, right? And then like then we might have to fix it. But this notion of like drift, and to your point, because the data that the model is training on is changing. The models are changing in some cases, and it's probabilistic. It's not a deterministic technology. So it's almost like a whole different lens that you need to put on the solutions that your teams are building. And I love your point about that ongoing maintenance. It's such a good question and like. What a big risk as a project manager, first of all, to be like, this thing is ready. Even though that might have actually been step one, you know, you go back to your project sponsor, your stakeholders, you're like, yeah, oh yeah, we're ready to launch. And them being like, oh, well, is there gonna be any like ongoing maintenance required? And you're like, Nope.

Kesha Williams:

I had it. We're done. Train the model. It's ready.

Galen Low:

It's good. It's working. It works great. It's gonna continue to work. Great. And we'll move on to the next phase. That I've seen that on not even just AI, but I've seen it kind of go wrong. I like the kind of the question you had asked, right? Like, okay, what does that mean in terms of accuracy? What's the next step? And I'd add to that just like what does this unlock and where are we at in the process? And even I was having this conversation with some members of my community the other day and I realized that like I'm that kind of person. Who will, not to toot my own horn, but like it's always a collaborative thing. It's like, okay, help me understand this because you know, I'm gonna have to go and explain this to my stakeholders. Is it okay if I say this? And then somebody like no. We cannot push the project. Do not tell them the fashion show is accurate. The team system models already, I don't know, like, you know, it's like asking that question to like collaboratively understand. And I think that's sometimes one of the big gaps is that, you know, I know a lot of folks. In a project management role, constantly frustrated about their teams.'cause they're like, oh, they don't get what I do. And I'm like, have you told them what you do and what your next step is? Because they're telling you what they know, right? It's like the model's already, the accuracy is good, and you're like, cool, thanks. And they're never gonna learn what you need to do with that. And then also they don't have an opportunity to intervene, to say actually, here's some things that are important that you need to tell the stakeholders. And please don't explain it this way. Please explain it this other way. I think that's like a really important dialogue to even just, like I said earlier, I'm like, well, PMs, we're used to deterministic actually, so is everybody else to even just build this like dialogue around and then it might drift. It's a very important thing to then go and educate stakeholders about, you know, it's not because we built it wrong, it's because this is the nature of the technology.

Kesha Williams:

That's the shift, that mindset shift now when we're building AI systems. And I really like your point about collaboration. Like that soft skill is even more important now in this age of AI.

Galen Low:

I love it. And also like tying back to your original point, which is like everyone needs to sort of build this foundational knowledge to have these conversations and it kind of needs to be collaborative because we're all figuring it out, you know?

Kesha Williams:

Right.

Galen Low:

There's not exactly, not enough time has passed and things are changing enough that there aren't really defacto experts or masters, I guess I should say. There's experts, but maybe not masters.

Kesha Williams:

Right.

Galen Low:

And we all need to kind of figure it out together. You gave some really good examples earlier if I wanted to kind of dig into like how someone, like a project manager, but frankly anyone who feels like they're not technical enough to really understand AI. What are your top tips for these folks who like need to act as some kind of translator between the teams? How does one go about. Learning enough to drive these conversations. Like what are maybe some of your go-to sources to learn that foundation, but also like stay up to speed as things are changing? I know you said don't chase all the tools, I get it. Don't like memorize all the release notes every time a new technology comes out. But where are some places you go or that you send your learners to stay up to speed and what are some like core foundational things that you might recommend as a, hey, you know, stakeholder x. You've only heard the word AI, but haven't scratched the surface beyond that. Read this thing.

Kesha Williams:

Right. And so I would have to point people to my LinkedIn learning courses and really more specifically that 15 day AI challenge because it really starts with the foundations and we start very simple, just helping you understand the different concepts, and then just really understanding how AI can plug into your day to help. It's not even about building systems, but it's how to use AI to boost your own productivity and help you in your day-to-day job. And so it's really important just to understand how AI can be applied. And as you start to do that and start to practice using it, you'll start to learn the concepts and then you'll be inspired. To come up with even like a brand new way for how you can use AI, how your team can use AI. So there are a lot of resources out there. I would say also follow my Kesha on AI LinkedIn page, and I also have a substack of that same name. So starting next week I'm going to post through Substack and on that LinkedIn page, just short AI tips. For anyone that's subscribed to that substack of mine, and so I'll start posting AI tips to help people just start to understand just the basic concepts and really trying to bring people along on the AI journey. So there are just a lot of other resources out there as well. I love following the Open AI Academy. They have a lot of webinars, free courses. That's where I do a lot of my learning on some of their newer resources. And then again, just LinkedIn learning. I have several courses on that platform. And then I have on Pluralsight some technical courses, but also non-technical courses about prompt engineering, about looking across your teams and understanding like what AI skills they need and how to upskill your teams in AI. So there are a lot of places out there where people can learn.

Galen Low:

I like your call out. Even it just kinda like there's a bit of an aha moment. Something I've been struggling with, which is from a project management standpoint, sometimes we're talking about teams that are using AI to build something. Sometimes we're talking about teams that are building something that has AI as part of its architecture. Sometimes we're talking about AI to doula project. Like in our day-to-day. And I think that's a really interesting sort of you just laid it out very succinctly, that like, yes, there are these different sort of ways and that comes down to some foundational stuff. My question, especially around the like open AI Academy and some others, it's easy to get pretty deep and like I work with, you know, cross-functional teams and sometimes I'm like, if I click on this. Am I even gonna understand it or am I gonna like try and learn too much? Is it gonna get to page three? And it's like, you know, I'm like deep into like Python code. Is there like any guidance you can give on helping folks? Maybe just like not get too deep understand that like, okay, this is probably at my level versus, okay, this seems to be for someone who's like actually coding.

Kesha Williams:

You're right. If you're at a place where it's asking you to spin up a Jupyter Notebook or write Python code, you've definitely gone too deep. But the way, like I structure a lot of my courses, I start with the concept first, and then I'll follow up with a hands-on activity. Sometimes if the hands-on activity is like too complex, you're writing Python code. Just watch the conceptual parts of it, make sure you understand the concepts because those concepts translate across really all of the different tools that you're using. So even if it is a very technical course, just watch the concept videos and understand the concepts and you can always skip that, like the deep, technical hands-on piece of it.

Galen Low:

I haven't heard anyone say that, but it makes so much sense that like don't not watch it if you're interested. There may be some things that are over your head. As long as you're kind of aware of that, there's still gonna be something you can take away from it. And honestly, I'm a huge fan of just like hearing the language. I would just sit down and like listen to my technical architect, talk to their team of developers, and I'd be like. Not all of this makes sense, but I'm picking up some words and I'm picking up some of the concepts, and I can at least feel a little bit more confident that I can ask the right questions. I might not be the master at any of it. I can at least have enough to ask a question that might be a dumb question, but also sort of opens that dialogue.

Kesha Williams:

There are no dumb questions.

Galen Low:

Okay. You go. Well, especially with AI, I guess, right? Where it's like things are changing so fast.

Kesha Williams:

Right? We're all learning. We are all learning, right.

Galen Low:

This might be controversial. I wonder if I can get your hot take on certifications and I, you know, I think that's a, it's a wide berth right now. I think there's some certifications coming out around AI that are more developer focused. There's definitely project management ones like from my friends at Cognilytica, which is now PMI Cognilytica, which is about like, yeah, if you're building, it's kind of around that sort. Building probabilistic systems and understanding where the data sources are from a project management lens. But you know, you're someone who you very deliberately, I think gone into like bite-sized hands-on learning. What is your take on certifications around AI and why someone might get them, and would it be worth it?

Kesha Williams:

I love certifications. I always tell people certifications could get you the interview, but they will not get you the job.'cause it's really important, like you mentioned, that hands-on piece, it's really important to be able to actually do what you're interviewing for. But certifications one, like I said, it will get you that interview. It gives you a solid enough foundation to understand the concepts and how everything fits together. But then for more hands on roles specifically, like more of the technical roles. It's important to have that hands-on piece, an online portfolio where you showcase the projects that you've actually developed using those technologies. But that's the advice that I always give when it comes to certifications.

Galen Low:

I really like that. Yeah, I think that makes a lot of sense. I like the idea of an online portfolio, and I know some of my listeners will be like, yeah, but like for developers, right? But to your point, day 15 of your 15 day challenge is build an agent. Would you recommend that even a project manager should have as part of their sort of portfolio, a few things that they've built or some demonstrative evidence that they use AI?

Kesha Williams:

In this day and age because of where AI is today and where it's headed, having that as a project manager will put you in front of other project managers that don't have that. It's really important when I think about where we're going to be a year from now, three years from now, five years from now, like I mentioned, AI is touching every industry, every role. It's really important for people to figure out. Where AI fits into their day-to-day and how they can use it to automate a lot of the repetitive task and use it to boost their productivity. If not, they are going to be left behind. And that's just the honest truth.

Galen Low:

It's a very fast freight train, right. That there's, even if we want it, it'd be very difficult to put this back in the box. You know what I mean?

Kesha Williams:

Yeah. That's not happening.

Galen Low:

Even from a developer standpoint, but I like that point that you made about, it'll get you to the interview, but it won't get you the job. If you were hiring a developer or maybe anybody really, what is the question you'd ask in the interview that would be kind of make or break to be like, yeah, this person actually knows AI, or this person probably just took a course and has this certificate, but. Maybe it doesn't have the chops that I'm looking for.

Kesha Williams:

Specifically for a developer, I would look to see how, or even if they're using AI, to help them write code and if they are specifically which tools they're using and specifically which part, and give me a real example, which part of the development lifecycle, where did they incorporate AI? A lot of, when I interview people, it's important for people to give real world examples and not just talk about the theory.

Galen Low:

Yeah. Even looking into the future, I think that even could apply to non-developers as well, right? It's like, okay, you've built a bunch of agents to help you manage projects better. Tell me about how you use them, I think is a very valid question that I would, yeah, I'm definitely stealing that, by the way.

Kesha Williams:

And which tool did you use and why did you decide to use that tool? Because there, when you're specifically around building agents, there are so many different tools out there. There's copilot where you don't even have to know how to code. There's Amazon Bedrock agents every platform now. Like Salesforce, every platform has a way to build agents, and so just understanding which tool like they used and why they chose that tool.

Galen Low:

It's a really good point. I wonder if we can, we started talking about it, so I thought this might be a good segue to just like look a bit into the future. I know we can't look that deep into the future. Everything's changing. It could go anywhere. AGI could show up tomorrow and then destroy humanity like Hollywood tells us.

Kesha Williams:

That's what science fiction says.

Galen Low:

Yes. Yeah. Or it could be like. Marginal improvements in AI notetakers for the next three years. I don't know, but to your point, right, like AI is being infused into administrative and repetitive tasks. That seems to be like.

Kesha Williams:

Right.

Galen Low:

The productivity boost is kind of the focus right now. It is manifesting as autonomous agents, right? To your point, there's a waste to build them. You can do 14 days of a challenge and on the 15th day you can build an agent, like it's not so far out of reach.

Kesha Williams:

Right.

Galen Low:

Do you think that this changes the role of the project manager? In the delivery of like technical AI native solutions. And if so, what's your hot tip for how project managers can prepare for the future of AI that might not exist yet, but will probably impact their roles?

Kesha Williams:

It all boils down to just understanding what AI is, the concepts and how you can apply it to your day-to-day to boost your productivity. I think there's a lot of fear across the industry about AI replacing roles. That's really not what we should be afraid of right now. The most important thing is not being left behind, figuring out how to use AI in your role to make you better, better than the next person. So that you can like help yourself, help your organization continue to push the needle forward.

Galen Low:

I love that and I love the themes there around the concepts. Some of the foundations are changing, I guess is the other thing. But I like your call out of like how you stay up to speed and also not necessarily chasing the tools down and not filling your brain with all of the specific tool knowledge, because this could go anywhere. We're looking at like, you know, beta max, VHS. We're looking at like cassette tape, bay track kind of stuff like. Some of this, there will be winners and losers on the big tech front in terms of adoption investments and all those other things. But if we understand the foundations, like those core concepts and how they're evolving in our jobs, then we can kind of extrapolate from there without necessarily getting too bogged down into the details. Kesha, thanks so much. This has been so much fun. Just for a little bit of extra holiday fun, do you have a question that you wanna ask me?

Kesha Williams:

Sure. We can flip it and I'll be the person interviewing. The question I have for you of course, is we've been talking about AI, so how are you using AI in your day-to-day? And if you're not using it, why?

Galen Low:

I am using it. I'm not a heavy user right now. I'm using it as like my thought partner. I'm very much a fan of the like Ironman, Jarvis kind of workflow, for example, especially like phrasing like on an email. I'm not first draft person as in like I don't use AI to do my first draft and then edit it. I'd rather write the first draft and then have my AI challenge it. So I'll be like, act as you know, a really tough project sponsor. You're a CTO. Here are the things you care about. Here's what our relationship looks like. Beat this up. Tell me where you might object to some of the content of this email. I think that's really a strong for me because my personality is like second guess until it's good. I'm that person. You know, this post online, and I don't consider myself a creative person necessarily, but there's this post that shows up in my feed sometimes. It's like a creative person, like it's never done. You could just get to the point where you can't look at it anymore. And I'm like, that's me at everything. Right? Emails, newsletter, subject lines, you know, like project plans. I just like, I'm staring at, I'm like beating it up. And I spent a lot of time just trying to figure out what someone might object to before. I think it's done. AI has been a really good partner for me there to be like, have you considered this? And I don't mind if it's like sometimes a bit sycophantic. I haven't like, gone through and been like, no, just like be mean to me now. But because I can, like, I can look through it. Like I'm not like, oh, I'm awesome, right? I'm great at this. Like, it's never my takeaway, right? From like a ChatGPT response. Yeah, that's how I've been using it so far. But I am, I do wanna dive into some of the autonomous stuff. That's kind of my next port of call. We've got some custom gpt that we've built, you know, to make our workflow stronger. I love that, you know, some of the specific tasks that are repetitive but still require judgment or a bit of, you know, interaction or interplay. Like, it's been really good for that. I'm excited to get into agents though, but you know, it's like the main thing that you started with, it's like, first I need to figure out where it's gonna help. Then I do have to have like my stuff in order, right? The data in order, the reference material. Like I need to be able to train this like I was training a person if I want it to be an effective agent. And I think that's kind of, it's funny, one of those things where it's like the technology's not the problem. It's like are we organized enough to train it and give it enough context to do its job? And that was true with humans before.

Kesha Williams:

I think that you raise is a very great point there when I think about like the role of the project manager. Soon you'll be managing agents alongside humans. So start to think about how you treat AI as like a part of the team. That's something that's coming as well.

Galen Low:

Yeah. I love that. We were jamming the other day, one of my other guests about what a performance review looks like and should we do one probably, right? Like, it's like, you know, we're, sometimes we give better feedback to our models than we do to our teammates, but I think the other point that you made, I think is really interesting is like, what does team culture look like? I think right now AI has a lot of capability, but the accountability still rests with humans. I mean, couldn't just like point a finger and be like, well, I blame this agent. This agent dropped the ball. Whereas frankly, I've been in team meetings where you know, someone points at their colleague and goes, yeah, well design, drop the ball whole, right? But it's like, I think there will be an interesting accountability mix and I think team culture, the way we build team culture will change. Some of the responsibilities actually fall to agents and automation and you know, other pieces of AI technology.

Kesha Williams:

Right. I agree. And that's where that AI governance part comes into play.

Galen Low:

I'll have to have you back when we're tackling that because in some ways it's sooner than I think, right? Like that's actually around the corner.

Kesha Williams:

Right.

Galen Low:

But maybe for today let's leave it there. Kesha, thank you so much for spending the time with me today. I love this conversation. I had so much fun. I actually learned a ton, and I love what you're doing. I think the format of what you do is really interesting. I'm excited about the Substack. For folks who are as excited as I am, where can they learn more about you?

Kesha Williams:

Definitely follow me on LinkedIn. Like you mentioned, I post there a lot. And then on my KeySoft company website, www.keysoft.tech, you can find more information about me and the services I provide on that website.

Galen Low:

Awesome. I will include those links in the show notes for folks who are interested. And Kesha, thanks again.

Kesha Williams:

Thank you. My pleasure. It was an awesome conversation.

Galen Low:

Alright folks, that's it for today's episode of The Digital Project Manager podcast. If you enjoyed this conversation, make sure to subscribe wherever you're listening. And if you want even more tactical insights, case studies and playbooks, head on over to thedigitalprojectmanager.com. Until next time, thanks for listening.