June 10, 2025 • 23 mins
In this episode of The Digital Project Manager podcast, Kelsey Alpaio sits down with Oliver Yarbrough—founder of Not Your Father’s AI and a LinkedIn Learning instructor—to talk about the AI-driven future of project management. With automation creeping into every corner of our workflows, PMs are no longer just task managers; they're becoming AI operators, process optimizers, and digital workforce leaders.
Oliver breaks down what PMs should be doing today to stay ahead: learning to communicate effectively with AI tools, mastering process optimization, and strengthening foundational PM skills. Whether you're just starting to explore AI tools or already dabbling in automation, this episode gives you a roadmap for evolving your skill set and redefining your value as a project manager in an increasingly digital workplace.
Resources from this episode:
- Join DPM Membership
- Subscribe to the newsletter to get our latest articles and podcasts
- Connect with Oliver on LinkedIn
- Check out Not Your Father's A.I.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Kelsey Alpaio (00:05):
Hey! I'm Kelsey.
And welcome back to The Retroon The Digital Project Manager
podcast—where we dig into pastlessons, future trends, and
what they mean for your career.
So today we're diving intoa topic that's pretty much
impossible to ignore rightnow, and a topic you really
shouldn't be ignoring—howAI will impact the future
of project management.
(00:26):
So at this point, we've allheard about the different tools,
all of the use cases for AI thatcan save you time and take those
annoying tasks off your plate.
But how much thought have yougiven to how AI will reshape
what it actually means to bea project manager—and what
skills you should be buildingnow to make sure you have a
long and successful careerin this field moving forward?
(00:49):
That's what we're gettinginto today, and we have a
great guest here to helpguide us—Oliver Yarbrough.
Oliver is the founder of NotYour Father's AI, which is all
about helping people apply AIin their jobs and businesses.
He's also a LinkedInLearning Instructor with a
popular course on artificialintelligence for project
managers, and he's a longtimeproject management trainer.
(01:13):
Oliver, welcome to the show!
Oliver Yarbrough (01:15):
Thanks.
I'm glad to be here.
Kelsey Alpaio (01:16):
Oliver, tell us a little bit more about yourself
and the work that you do.
Oliver Yarbrough (01:20):
I'm a project management trainer first and
foremost, and in the lasteight years or so, I started
talking about the artificialintelligence specifically as it
relates to project management.
As of, the last couple ofyears I've been doing this
thing called Not Your Father'sAI, helping people just
understand the practical usesof artificial intelligence.
Not just for project management,for other areas as well, but for
(01:41):
purposes of our conversation,of course, project management.
Kelsey Alpaio (01:45):
Great.
So let's get right into ithere and talk a little bit
about how AI is alreadystarting to reshape the
project management role.
So what are some of the majorshifts you're already seeing?
Oliver Yarbrough (01:55):
I'm seeing people take what they do on a
daily basis and starting to justincorporate it at a small level.
So of course there are peoplewho are using it at a much
larger level, and we'lltalk about that as well.
But it's basically findingout, what you're doing and
figuring out how can I squeezeinefficiencies out of those
by hopefully incorporatingvarious aspects of artificial
(02:18):
intelligence and or automation.
Those are some of the basicareas I would say there.
Then I'm also seeingpeople use it to just
help them write better.
A lot of people, even if we'regood writers some of us want
to enhance it, or at leastnot start with a blank page.
So I don't care if it's apresentation, if it's, an email.
(02:41):
You never have to reallystart with a blank page.
So those are just some of thebasic areas I'm seeing there.
Kelsey Alpaio (02:46):
And how do you see AI continuing to
make an impact over thenext five years or so?
Like what is the PM rolereally going to look like
as AI continues to make itsmark on this profession?
Oliver Yarbrough (02:58):
That's a very good question.
And literally this morning Ihad a call with another company
and where I see this evolvingto, it's, unlike some people
who have this no, no scaretactics and stuff, it's not
really gonna be the eliminationof the project management role.
What's going to happen isyou're gonna have the people who
embrace it, who really learn it.
(03:21):
Are going to be more or lessoperators, they're going
to be the ones directing,training and retooling.
But literally, and I'mnot talking about this
happening like next week,but possibly over the next
2, 3, 4, 5 years, they'regonna be managing a group of
digital stakeholders and oremployees and or coordinators.
(03:41):
And so a lot of these things,it's gonna be, project
managers are gonna be moreoperators, is what I'm saying.
And you're gonna have tounderstand project management.
Then also understand how tolead, and this is gonna be
a whole new field here, howto lead digital agents, just
like human beings, whatever.
So in some environments it'sgonna be more like a hybrid
environment where there aregonna be human coordinators
(04:04):
working underneath, likea project manager or some
other new title, anyway.
And so also a digital agent too.
And so I call that moreof a hybrid environment.
Kelsey Alpa (04:13):
Really interesting.
So I wanna dig in a little bitmore to what those skills are
that project managers shouldbe building right now to help
prepare themselves for that.
So first off, how shouldPMs be thinking about
AI on the day-to-day?
What kind of mindset should theyhave around using AI in their
current roles that can helpprepare them for that future?
Oliver Yarbrough (04:34):
I think that it really boils
down to three core areas.
I'm not saying thatthese are the only areas.
We've all heard about softskills and all these things,
and those are very important.
Okay.
In this new era, I thinkit really narrows down
to three core areas.
Number one isproject management.
You need to know your field.
You need to knowproject management.
It's hard to evaluate humanbeings or agents if you don't
(04:55):
understand project management.
So that's number one.
Number two is it's understandingartificial intelligence.
I'm not saying that you haveto go out and learn python and
no programming, but you need tounderstand how these LLMs think.
And that's, like that'snot super, super technical.
I'm not saying again,that you need to go learn
how to program it, butunderstand how it thinks.
And then number three, whichis not necessarily on a lot
(05:17):
of people's radar, but Ithink it's just as important,
is you need to understandprocess optimization.
How to optimize a process.
And why do I say this?
Because in order to get themost use out of AI, you're
gonna have to figure out howto squeeze out inefficiencies
in your processes.
That entails at a basic level,let's just say in any given
(05:39):
week you do a hundred tasks,is figuring out of those a
hundred tasks, which ones haveto be done by a human being.
And then of those onesthat have to be done by
a human being, which oneshave to be done by me, or
which ones can I delegate?
And then of those other onesthat can't be done by a human,
which ones can be done by AI?
(06:02):
Then of course the thirdcategory in there is which
ones should not be done at all.
The combination of thosethree, eliminating ones
which should not be done,outsourcing tasks to either
AI or to other human beings.
And then what's left forme as the project manager
to ultimately do, I cannotoutsource it to anybody
else or anything else.
I think that's whereI really focus my
(06:25):
attention going forward.
Kelsey Alpaio (06:28):
Got it.
So can you talk a little bitabout the process of developing
those skills now for the future?
So are there specific AI skills,PMs should be laser focused
on building, and what are thethings that are going to be deal
breakers if you don't know howto do them in the near future?
Is it prompt building,is it something else?
Let's dig in therea little bit more.
Oliver Yarbrough (06:45):
Okay.
So we're talking aboutwhat we typically call AI.
Okay.
What most people call AI.
If we start and I have thisconcept that I've talked
about for a long time.
It's project stakeholders,AI stakeholders.
If we're gonna treat AI asif it's a stakeholder, as if
it's any other stakeholder,if you're talking about
like a RACI diagram, right?
Like basically it's astakeholder, it's one of the
(07:07):
stakeholders on your projects.
You have to know, like anyother stakeholder, you have
to know how to engage it,how to interact with it.
So it's really finding outwhat that agent or what that
AI bot needs in order togive you the optimal answers.
So how do wecommunicate with AI?
(07:28):
The basic level is promptengineering, so the prompt
engineering title might go away,but the need to prompt or write
good prompts is not going away.
This is basically knowinghow to engage your bot.
So I would say, promptengineering is very important,
more or less knowing, how towrite a good prompt and there's
(07:50):
different levels to it, right?
As I stated earlier, knowinghow to optimize a process.
So take you like some ofSix Sigma or something.
Take any classes or coursesthat will help you streamline
a process and squeeze outany type of inefficiencies.
And then as a third level, Iwould definitely say some type
of soft skills, meaning that youknow how to, negotiate, you know
(08:13):
how to buy things and variousthings that compliment what
your AI agent can't do or maybecan't do as well as you like.
Anything that involveshuman engagement.
Here's why.
It will allow you tocompliment what AI does.
You'll actually add valuein a different sense.
You'll add value to whatAI is already doing.
(08:36):
I think that those are whereI would spend more time.
Kelsey Alpaio (08:38):
That's great.
I wanna do a little bitof a, like 101, 201, 301
here, for those who mightbe wondering, where to
start and how to level up.
So at the 101 level, forsomeone who's maybe a little
newer to AI or try to, it waslike, eh, what advice do you
have to help them get startedon building these skills?
Oliver Yarbrough (09:00):
That's a great question because so many
people get overwhelmed becausethey come in and they, oh, I
need to start to build chartersand I need to start to do
sprint planning and everything.
Start with something simple.
Start with how do Istreamline and automate
taking meeting minutes?
How do I create apresentation that I can
then deliver to my team?
(09:22):
Why do I say this?
Because even though this mightnot be the technical aspects
of managing projects, what itis it gets you a quick win.
It gets you feelingsome of the time saved.
Initially it'sabout saving time.
So yes, I could sit hereand take meeting minutes.
I could sit here and take anhour and create a PowerPoint
presentation, or I could writea prompt, load up a document
(09:46):
or two and have a presentationdone in five minutes or less.
Now, maybe I do need toobviously tweak it a bit, but
I don't start from scratch.
For example, I cantake meeting minutes.
I can then load thoseminutes into if I don't
wanna do full on automation.
I can take those meetingminutes, I can then load them
into a chat bot and summarizeit, and then I could then
(10:08):
send it out as a summary.
That's if I use no automation.
If I use automation, itgets a whole lot better.
Kelsey Alpaio (10:15):
Yeah, that's a great, I feel like first step.
And so at that 201 level,once you feel like you've
got a handle on it, you'vebeen using it day to day
for a bit, you're in agroove, what's the next step?
What's that surprising use caseyou might not have thought of?
How can you advance yourskills from the basics and
really start to do some of thatfuture proofing of yourself?
Oliver Yarbrough (10:36):
So at the 201 level, I'm taking it from the
basics, I just talked about.
Now we're gonna do somethinglike build our own custom GPT.
This would require apaid account, and every
type of LLM has liketheir own version of it.
Gyms and Google, other oneshave other ones, right?
So what I wanna do now is in apaid account, either in a, for
(10:57):
example, ChatGPT teams account.
It comes native, right?
It comes where you don't haveto worry about your data being
shared with the outside world.
But if we use a regular $20a month account, you do need
to designate, don't share.
So I'm saying this so that youfeel comfortable now uploading
your personal information to it.
We can train a GPT onour specific information.
(11:21):
And here's where I think alot of people are challenged.
They try to have oneGPT do everything.
Like they load it with 50.
Documents and justhave it do everything.
No.
What I wanna encouragepeople to do is to train
a GPT to do one specifictask and do it very well.
So for example, I want youto create project charters.
(11:41):
All this does is itcreates project charters.
It doesn't do every single bank.
All it does is I load itwith examples of great
project charters andother information I want.
I then put instructions in it,which for example, you are a
project manager working forX, Y, Z company with 15 years
(12:03):
experience and other details.
So now the reason I dothat in the instructions
is now I don't have to typethat every single time.
So now when I come in and writea prompt for this new charter,
all I have to do is say, writeme a charter for X, Y, Z.
It gets pretty goodif it's fine tuned.
(12:24):
The train rate.
So that would, I wouldsay, would be the easiest
and probably best wayto go in at a 201 level.
Kelsey Alpaio (12:31):
That's great.
Yeah.
So at the 301 level, likewe're talking basically expert
users here, what are someof the skills and tools to
be focused on at that level?
What does the ultimate projectmanager, AI user look like?
Oliver Yarbrough (12:45):
I would say where we are now, you're
getting into automation.
At this level, you're gettinginto intelligent automation.
This is where you're not justusing a chat bot, but you're
connecting them and you're usingsoftware such as a make.com.
Make.com, Zapier, and what thesedo is these integrate these
(13:07):
allow you to pull in informationfrom one piece of software and
bringing it to like another one.
So for example, here,let's just say like you're
using a software like yourproject management software.
Something like Clickupor whatever, right?
You can, for example, use anAirtable or a Google Sheet
as a database of sorts.
(13:27):
You can pull information fromClickup into an Airtable.
You can then, for example,take that information from the
Airtable and run it through,say like a ChatGPT, bring
it back into the Airtable,and then spin it out into
a dashboard or something.
This is also, of course,assuming you're not using the
AI that's already built intovarious pieces of software,
(13:50):
which may already do a lotof these things anyway.
But one of the reasons you maywanna have your own database
is in case you want to havea record of it in your own
environment, and you may wantto do other things with it
outside of just managing likeone or two projects for example.
So I would say, to answeryour question, in a nutshell,
it's intelligent automation.
Kelsey Alpaio (14:12):
Great.
You mentioned this a little bitwith a couple of those stages
there, but one challenge I hearabout often is PM saying I'm
really excited about using AIand I really wanna be using
it more regularly for mywork, but my organization's
just not there yet.
They're wary of it.
They're not sure of thesecurity implications.
So what advice do you havefor someone in that position
(14:34):
who wants to be buildingthese skills but is limited
by their organizationor access to the tools?
Oliver Yarbrough (14:41):
I would encourage people, whenever
I'm faced with pushback ora challenge, and I've had
this back when I worked incorporate America, doesn't
matter what job you're in,you're gonna face pushback.
The first thing I ask myselfis, why am I getting pushback?
What's the root cause?
What's driving this?
What's the fear factor here?
Because without understandingwhat's the fear factor or
(15:02):
what's driving it, it'sreally hard to come back
and offer like a reason.
For example, if the questionthey're just asking, we
find out it's a privacyissue I can address that.
I can let them know that,for example, if you are
using the Microsoft Suite,they have privacy built in.
It's not gonna be used totrain the larger data set.
(15:22):
If we're already usingMicrosoft products already,
then it's gonna be lessof a less of a hard sell.
If we're using ChatGPT,we can have a paid account
or a teams account.
For example, if we'reusing a teams account in
ChatGPT, it comes natively.
It does not share informationwith the larger dataset.
(15:43):
So if privacy is the maindriver for not using it,
then that sort of letssome of that fear go away.
If it's just the fear that if AIis going to just take our jobs,
then there's other ways in whichyou can address that as well.
By, again, as I statedearlier, showing them
how they can save time.
Showing them how it'snot necessarily going to
take their jobs as muchas it's going to enhance
(16:04):
what they're already doing.
And so over time we canhopefully start to chip
away at that fear, thathesitancy and get people
to ultimately move forward.
Kelsey Alpaio (16:14):
Yeah, that's great advice.
I do feel like that aspectof fear, kind of a fear that
AI is going take your job isone of the challenges that I
think some people run into.
And I'm wondering if there areother, common traps that you
see PMs falling into when theystart building these skills.
Are there misuses, overusesjust bad habits, mindsets
(16:36):
that they need to overcome?
Can we talk a littlebit about that?
Oliver Yarbrough (16:39):
Yeah, I could talk about this
for like days, right?
But some of the ones that cometo mind are relying too heavily
on the output that you get.
AI is not perfect.
Humans are not perfect.
If humans aren't perfect,the things we program
won't be perfect.
And why do I say this?
Because you may have heardthis term, hallucinations.
(17:00):
AI has a tendency at timesto hallucinate, to make up
stuff, to make you feel good.
It doesn't wannamake you feel bad.
If it doesn't have an answer,sometimes it will just give you
what it thinks you want to hear.
So there is some precedencein prompting to give
your AI permission tonot get everything right.
(17:24):
Literally prompt it, give mean answer, X, Y, Z. But if
you don't know the answer,tell me you don't know it.
Do not hallucinate.
Do not give me, donot make stuff up.
Tell it that.
So that's one thing I wouldsay is something that I would
definitely not overlook also.
You have to continue to improveyour project management skills.
(17:46):
I cannot stress this enough,whether it be a traditional,
hybrid, agile, whatever themethodology is, you have
to be well versed to thatbecause, and this is especially
important, I'm really sayingthis not for the more seasoned
people, but for the people whoare maybe transitioning into
the field from other areasor coming out of college.
(18:08):
You need to really getcertified, really learn the
various aspects of projectmanagement because that will
ultimately make you a betteroperator down the road when
it comes to building andmanaging your digital workforce.
Kelsey Alpaio (18:24):
Absolutely.
I think another thing too isthat I get tripped up on is
this fear of can I keep up?
AI changes so fast and thereare new tools popping up every
week, new versions of thingsgetting released like every day.
And with so much changingso fast, how should we be
thinking about staying currentwith AI and with these skills
(18:47):
without it becoming like yourfull-time job, basically or
should it be your full-time job?
Oliver Yarbrough (18:52):
I love this question.
I'm gonna give you like a 101answer and like a 201 answer.
Okay?
The 101 answer is listen topodcasts like yours, watch
YouTube videos like yours.
Read blogs and newsletters.
I have an email account setup just to get newsletters and
blog posts and stuff come inand I have filters in it, which
(19:13):
I mean like all kind of o otherstuff there, but literally.
Okay, so that's 101.
It's like just go out,read, listen, digest as
much stuff as possible, takecourses, everything else.
The 201 level is automated.
For example, have sign upfor these newsletters, these
blog posts, have them comeinto a dedicated email.
(19:35):
Say for example, like a Gmail,have filters in there, and
then you can have a bot go inthere, pull those out, summarize
them, find the key points,put them in a daily reminder
that summarizes it in bulletpoints every single day from
multiple emails and newslettersand even YouTube videos.
You can get this summarized.
(19:55):
So it's basically likehaving your own personal
assistant summarize it andgive it to you in bullet
points like once a day.
It's almost like having yourown daily AI standup that's
been summarized in advance.
Kelsey Alpai (20:08):
Yeah, I love that.
A way to make the learninga part of your day, but even
just using AI to make thatprocess easier is great.
I love that.
Oliver Yarbrough (20:16):
Yeah.
And the beautiful thing isif you're already in Gmail,
a Gemini is built into it.
Mm-hmm.
So the, the beautiful thingabout having AI built into
something is you don't have touse automation anymore, really.
It's already there.
So it can summarizeit instantly.
Kelsey Alpaio (20:31):
That's great.
So we're just about out oftime here, but to wrap things
up for us, can you leave uswith one thing every listener
can do today to improvetheir AI skills and start to
future proof their PM career?
Oliver Yarbrough (20:44):
One thing that you can do to future proof
your career, I would say iseven though it's getting to be
old school now, it's been, it'sbeen out two or three years,
good old prompt engineering.
Okay.
And not just likeyour standard one.
Okay.
I would say that evenif you're a newbie.
You should probably, honestlyspeaking, start to learn custom
(21:06):
GPTs, even though it's a 201level, the basic level is gonna
be, it's gonna become so basic.
It's gonna be like talkingabout the internet like in
2025, like who talks abouthow to get online in 2025.
So I'm gonna encourageyou to advance from the
101 level to the 201 levelas quickly as possible.
(21:28):
Start learning customGPTs and more specifically
how to prompt for them.
When I say this, specificallyhow to write instructions.
Instructions, meaning thatyou're telling it you are
X, Y, Z, working in thissize company that becomes
basically your avatar.
You're basically, you'recreating a digital avatar
(21:49):
that you don't have to thenprompt it every single time.
That's the first level there,and then learning how to
write the prompt after that.
That's more of a workshopthere, but if you just
spend a little bit of time,take a couple of courses,
a couple of videos, learnthat it will do you wonders.
It will be a tool in yourtool belt that will give you
(22:10):
a superpower for the next 18months until everyone else
jumps on board and then itbecomes like a common knowledge.
Kelsey Alpaio (22:16):
Yeah.
I love that.
Oliver, thank you so much forspending time with me today.
This has been so helpful.
Oliver Yarbrough (22:21):
Thank you.
I appreciate it.
Kelsey Alpaio (22:28):
That's it for today's Retro.
Be sure to follow the showso you never miss an episode.
And if you wanna keep theconversation going with a crew
of 1,000+ project managementpros who get it, come join
us at thedpm.com/membership.
Thanks for listening!
Hey! I'm Kelsey.
And welcome back to The Retroon The Digital Project Manager
podcast—where we dig into pastlessons, future trends, and
what they mean for your career.
So today we're diving intoa topic that's pretty much
impossible to ignore rightnow, and a topic you really
shouldn't be ignoring—howAI will impact the future
of project management.
(00:26):
So at this point, we've allheard about the different tools,
all of the use cases for AI thatcan save you time and take those
annoying tasks off your plate.
But how much thought have yougiven to how AI will reshape
what it actually means to bea project manager—and what
skills you should be buildingnow to make sure you have a
long and successful careerin this field moving forward?
(00:49):
That's what we're gettinginto today, and we have a
great guest here to helpguide us—Oliver Yarbrough.
Oliver is the founder of NotYour Father's AI, which is all
about helping people apply AIin their jobs and businesses.
He's also a LinkedInLearning Instructor with a
popular course on artificialintelligence for project
managers, and he's a longtimeproject management trainer.
(01:13):
Oliver, welcome to the show!
Oliver Yarbrough (01:15):
Thanks.
I'm glad to be here.
Kelsey Alpaio (01:16):
Oliver, tell us a little bit more about yourself
and the work that you do.
Oliver Yarbrough (01:20):
I'm a project management trainer first and
foremost, and in the lasteight years or so, I started
talking about the artificialintelligence specifically as it
relates to project management.
As of, the last couple ofyears I've been doing this
thing called Not Your Father'sAI, helping people just
understand the practical usesof artificial intelligence.
Not just for project management,for other areas as well, but for
(01:41):
purposes of our conversation,of course, project management.
Kelsey Alpaio (01:45):
Great.
So let's get right into ithere and talk a little bit
about how AI is alreadystarting to reshape the
project management role.
So what are some of the majorshifts you're already seeing?
Oliver Yarbrough (01:55):
I'm seeing people take what they do on a
daily basis and starting to justincorporate it at a small level.
So of course there are peoplewho are using it at a much
larger level, and we'lltalk about that as well.
But it's basically findingout, what you're doing and
figuring out how can I squeezeinefficiencies out of those
by hopefully incorporatingvarious aspects of artificial
(02:18):
intelligence and or automation.
Those are some of the basicareas I would say there.
Then I'm also seeingpeople use it to just
help them write better.
A lot of people, even if we'regood writers some of us want
to enhance it, or at leastnot start with a blank page.
So I don't care if it's apresentation, if it's, an email.
(02:41):
You never have to reallystart with a blank page.
So those are just some of thebasic areas I'm seeing there.
Kelsey Alpaio (02:46):
And how do you see AI continuing to
make an impact over thenext five years or so?
Like what is the PM rolereally going to look like
as AI continues to make itsmark on this profession?
Oliver Yarbrough (02:58):
That's a very good question.
And literally this morning Ihad a call with another company
and where I see this evolvingto, it's, unlike some people
who have this no, no scaretactics and stuff, it's not
really gonna be the eliminationof the project management role.
What's going to happen isyou're gonna have the people who
embrace it, who really learn it.
(03:21):
Are going to be more or lessoperators, they're going
to be the ones directing,training and retooling.
But literally, and I'mnot talking about this
happening like next week,but possibly over the next
2, 3, 4, 5 years, they'regonna be managing a group of
digital stakeholders and oremployees and or coordinators.
(03:41):
And so a lot of these things,it's gonna be, project
managers are gonna be moreoperators, is what I'm saying.
And you're gonna have tounderstand project management.
Then also understand how tolead, and this is gonna be
a whole new field here, howto lead digital agents, just
like human beings, whatever.
So in some environments it'sgonna be more like a hybrid
environment where there aregonna be human coordinators
(04:04):
working underneath, likea project manager or some
other new title, anyway.
And so also a digital agent too.
And so I call that moreof a hybrid environment.
Kelsey Alpa (04:13):
Really interesting.
So I wanna dig in a little bitmore to what those skills are
that project managers shouldbe building right now to help
prepare themselves for that.
So first off, how shouldPMs be thinking about
AI on the day-to-day?
What kind of mindset should theyhave around using AI in their
current roles that can helpprepare them for that future?
Oliver Yarbrough (04:34):
I think that it really boils
down to three core areas.
I'm not saying thatthese are the only areas.
We've all heard about softskills and all these things,
and those are very important.
Okay.
In this new era, I thinkit really narrows down
to three core areas.
Number one isproject management.
You need to know your field.
You need to knowproject management.
It's hard to evaluate humanbeings or agents if you don't
(04:55):
understand project management.
So that's number one.
Number two is it's understandingartificial intelligence.
I'm not saying that you haveto go out and learn python and
no programming, but you need tounderstand how these LLMs think.
And that's, like that'snot super, super technical.
I'm not saying again,that you need to go learn
how to program it, butunderstand how it thinks.
And then number three, whichis not necessarily on a lot
(05:17):
of people's radar, but Ithink it's just as important,
is you need to understandprocess optimization.
How to optimize a process.
And why do I say this?
Because in order to get themost use out of AI, you're
gonna have to figure out howto squeeze out inefficiencies
in your processes.
That entails at a basic level,let's just say in any given
(05:39):
week you do a hundred tasks,is figuring out of those a
hundred tasks, which ones haveto be done by a human being.
And then of those onesthat have to be done by
a human being, which oneshave to be done by me, or
which ones can I delegate?
And then of those other onesthat can't be done by a human,
which ones can be done by AI?
(06:02):
Then of course the thirdcategory in there is which
ones should not be done at all.
The combination of thosethree, eliminating ones
which should not be done,outsourcing tasks to either
AI or to other human beings.
And then what's left forme as the project manager
to ultimately do, I cannotoutsource it to anybody
else or anything else.
I think that's whereI really focus my
(06:25):
attention going forward.
Kelsey Alpaio (06:28):
Got it.
So can you talk a little bitabout the process of developing
those skills now for the future?
So are there specific AI skills,PMs should be laser focused
on building, and what are thethings that are going to be deal
breakers if you don't know howto do them in the near future?
Is it prompt building,is it something else?
Let's dig in therea little bit more.
Oliver Yarbrough (06:45):
Okay.
So we're talking aboutwhat we typically call AI.
Okay.
What most people call AI.
If we start and I have thisconcept that I've talked
about for a long time.
It's project stakeholders,AI stakeholders.
If we're gonna treat AI asif it's a stakeholder, as if
it's any other stakeholder,if you're talking about
like a RACI diagram, right?
Like basically it's astakeholder, it's one of the
(07:07):
stakeholders on your projects.
You have to know, like anyother stakeholder, you have
to know how to engage it,how to interact with it.
So it's really finding outwhat that agent or what that
AI bot needs in order togive you the optimal answers.
So how do wecommunicate with AI?
(07:28):
The basic level is promptengineering, so the prompt
engineering title might go away,but the need to prompt or write
good prompts is not going away.
This is basically knowinghow to engage your bot.
So I would say, promptengineering is very important,
more or less knowing, how towrite a good prompt and there's
(07:50):
different levels to it, right?
As I stated earlier, knowinghow to optimize a process.
So take you like some ofSix Sigma or something.
Take any classes or coursesthat will help you streamline
a process and squeeze outany type of inefficiencies.
And then as a third level, Iwould definitely say some type
of soft skills, meaning that youknow how to, negotiate, you know
(08:13):
how to buy things and variousthings that compliment what
your AI agent can't do or maybecan't do as well as you like.
Anything that involveshuman engagement.
Here's why.
It will allow you tocompliment what AI does.
You'll actually add valuein a different sense.
You'll add value to whatAI is already doing.
(08:36):
I think that those are whereI would spend more time.
Kelsey Alpaio (08:38):
That's great.
I wanna do a little bitof a, like 101, 201, 301
here, for those who mightbe wondering, where to
start and how to level up.
So at the 101 level, forsomeone who's maybe a little
newer to AI or try to, it waslike, eh, what advice do you
have to help them get startedon building these skills?
Oliver Yarbrough (09:00):
That's a great question because so many
people get overwhelmed becausethey come in and they, oh, I
need to start to build chartersand I need to start to do
sprint planning and everything.
Start with something simple.
Start with how do Istreamline and automate
taking meeting minutes?
How do I create apresentation that I can
then deliver to my team?
(09:22):
Why do I say this?
Because even though this mightnot be the technical aspects
of managing projects, what itis it gets you a quick win.
It gets you feelingsome of the time saved.
Initially it'sabout saving time.
So yes, I could sit hereand take meeting minutes.
I could sit here and take anhour and create a PowerPoint
presentation, or I could writea prompt, load up a document
(09:46):
or two and have a presentationdone in five minutes or less.
Now, maybe I do need toobviously tweak it a bit, but
I don't start from scratch.
For example, I cantake meeting minutes.
I can then load thoseminutes into if I don't
wanna do full on automation.
I can take those meetingminutes, I can then load them
into a chat bot and summarizeit, and then I could then
(10:08):
send it out as a summary.
That's if I use no automation.
If I use automation, itgets a whole lot better.
Kelsey Alpaio (10:15):
Yeah, that's a great, I feel like first step.
And so at that 201 level,once you feel like you've
got a handle on it, you'vebeen using it day to day
for a bit, you're in agroove, what's the next step?
What's that surprising use caseyou might not have thought of?
How can you advance yourskills from the basics and
really start to do some of thatfuture proofing of yourself?
Oliver Yarbrough (10:36):
So at the 201 level, I'm taking it from the
basics, I just talked about.
Now we're gonna do somethinglike build our own custom GPT.
This would require apaid account, and every
type of LLM has liketheir own version of it.
Gyms and Google, other oneshave other ones, right?
So what I wanna do now is in apaid account, either in a, for
(10:57):
example, ChatGPT teams account.
It comes native, right?
It comes where you don't haveto worry about your data being
shared with the outside world.
But if we use a regular $20a month account, you do need
to designate, don't share.
So I'm saying this so that youfeel comfortable now uploading
your personal information to it.
We can train a GPT onour specific information.
(11:21):
And here's where I think alot of people are challenged.
They try to have oneGPT do everything.
Like they load it with 50.
Documents and justhave it do everything.
No.
What I wanna encouragepeople to do is to train
a GPT to do one specifictask and do it very well.
So for example, I want youto create project charters.
(11:41):
All this does is itcreates project charters.
It doesn't do every single bank.
All it does is I load itwith examples of great
project charters andother information I want.
I then put instructions in it,which for example, you are a
project manager working forX, Y, Z company with 15 years
(12:03):
experience and other details.
So now the reason I dothat in the instructions
is now I don't have to typethat every single time.
So now when I come in and writea prompt for this new charter,
all I have to do is say, writeme a charter for X, Y, Z.
It gets pretty goodif it's fine tuned.
(12:24):
The train rate.
So that would, I wouldsay, would be the easiest
and probably best wayto go in at a 201 level.
Kelsey Alpaio (12:31):
That's great.
Yeah.
So at the 301 level, likewe're talking basically expert
users here, what are someof the skills and tools to
be focused on at that level?
What does the ultimate projectmanager, AI user look like?
Oliver Yarbrough (12:45):
I would say where we are now, you're
getting into automation.
At this level, you're gettinginto intelligent automation.
This is where you're not justusing a chat bot, but you're
connecting them and you're usingsoftware such as a make.com.
Make.com, Zapier, and what thesedo is these integrate these
(13:07):
allow you to pull in informationfrom one piece of software and
bringing it to like another one.
So for example, here,let's just say like you're
using a software like yourproject management software.
Something like Clickupor whatever, right?
You can, for example, use anAirtable or a Google Sheet
as a database of sorts.
(13:27):
You can pull information fromClickup into an Airtable.
You can then, for example,take that information from the
Airtable and run it through,say like a ChatGPT, bring
it back into the Airtable,and then spin it out into
a dashboard or something.
This is also, of course,assuming you're not using the
AI that's already built intovarious pieces of software,
(13:50):
which may already do a lotof these things anyway.
But one of the reasons you maywanna have your own database
is in case you want to havea record of it in your own
environment, and you may wantto do other things with it
outside of just managing likeone or two projects for example.
So I would say, to answeryour question, in a nutshell,
it's intelligent automation.
Kelsey Alpaio (14:12):
Great.
You mentioned this a little bitwith a couple of those stages
there, but one challenge I hearabout often is PM saying I'm
really excited about using AIand I really wanna be using
it more regularly for mywork, but my organization's
just not there yet.
They're wary of it.
They're not sure of thesecurity implications.
So what advice do you havefor someone in that position
(14:34):
who wants to be buildingthese skills but is limited
by their organizationor access to the tools?
Oliver Yarbrough (14:41):
I would encourage people, whenever
I'm faced with pushback ora challenge, and I've had
this back when I worked incorporate America, doesn't
matter what job you're in,you're gonna face pushback.
The first thing I ask myselfis, why am I getting pushback?
What's the root cause?
What's driving this?
What's the fear factor here?
Because without understandingwhat's the fear factor or
(15:02):
what's driving it, it'sreally hard to come back
and offer like a reason.
For example, if the questionthey're just asking, we
find out it's a privacyissue I can address that.
I can let them know that,for example, if you are
using the Microsoft Suite,they have privacy built in.
It's not gonna be used totrain the larger data set.
(15:22):
If we're already usingMicrosoft products already,
then it's gonna be lessof a less of a hard sell.
If we're using ChatGPT,we can have a paid account
or a teams account.
For example, if we'reusing a teams account in
ChatGPT, it comes natively.
It does not share informationwith the larger dataset.
(15:43):
So if privacy is the maindriver for not using it,
then that sort of letssome of that fear go away.
If it's just the fear that if AIis going to just take our jobs,
then there's other ways in whichyou can address that as well.
By, again, as I statedearlier, showing them
how they can save time.
Showing them how it'snot necessarily going to
take their jobs as muchas it's going to enhance
(16:04):
what they're already doing.
And so over time we canhopefully start to chip
away at that fear, thathesitancy and get people
to ultimately move forward.
Kelsey Alpaio (16:14):
Yeah, that's great advice.
I do feel like that aspectof fear, kind of a fear that
AI is going take your job isone of the challenges that I
think some people run into.
And I'm wondering if there areother, common traps that you
see PMs falling into when theystart building these skills.
Are there misuses, overusesjust bad habits, mindsets
(16:36):
that they need to overcome?
Can we talk a littlebit about that?
Oliver Yarbrough (16:39):
Yeah, I could talk about this
for like days, right?
But some of the ones that cometo mind are relying too heavily
on the output that you get.
AI is not perfect.
Humans are not perfect.
If humans aren't perfect,the things we program
won't be perfect.
And why do I say this?
Because you may have heardthis term, hallucinations.
(17:00):
AI has a tendency at timesto hallucinate, to make up
stuff, to make you feel good.
It doesn't wannamake you feel bad.
If it doesn't have an answer,sometimes it will just give you
what it thinks you want to hear.
So there is some precedencein prompting to give
your AI permission tonot get everything right.
(17:24):
Literally prompt it, give mean answer, X, Y, Z. But if
you don't know the answer,tell me you don't know it.
Do not hallucinate.
Do not give me, donot make stuff up.
Tell it that.
So that's one thing I wouldsay is something that I would
definitely not overlook also.
You have to continue to improveyour project management skills.
(17:46):
I cannot stress this enough,whether it be a traditional,
hybrid, agile, whatever themethodology is, you have
to be well versed to thatbecause, and this is especially
important, I'm really sayingthis not for the more seasoned
people, but for the people whoare maybe transitioning into
the field from other areasor coming out of college.
(18:08):
You need to really getcertified, really learn the
various aspects of projectmanagement because that will
ultimately make you a betteroperator down the road when
it comes to building andmanaging your digital workforce.
Kelsey Alpaio (18:24):
Absolutely.
I think another thing too isthat I get tripped up on is
this fear of can I keep up?
AI changes so fast and thereare new tools popping up every
week, new versions of thingsgetting released like every day.
And with so much changingso fast, how should we be
thinking about staying currentwith AI and with these skills
(18:47):
without it becoming like yourfull-time job, basically or
should it be your full-time job?
Oliver Yarbrough (18:52):
I love this question.
I'm gonna give you like a 101answer and like a 201 answer.
Okay?
The 101 answer is listen topodcasts like yours, watch
YouTube videos like yours.
Read blogs and newsletters.
I have an email account setup just to get newsletters and
blog posts and stuff come inand I have filters in it, which
(19:13):
I mean like all kind of o otherstuff there, but literally.
Okay, so that's 101.
It's like just go out,read, listen, digest as
much stuff as possible, takecourses, everything else.
The 201 level is automated.
For example, have sign upfor these newsletters, these
blog posts, have them comeinto a dedicated email.
(19:35):
Say for example, like a Gmail,have filters in there, and
then you can have a bot go inthere, pull those out, summarize
them, find the key points,put them in a daily reminder
that summarizes it in bulletpoints every single day from
multiple emails and newslettersand even YouTube videos.
You can get this summarized.
(19:55):
So it's basically likehaving your own personal
assistant summarize it andgive it to you in bullet
points like once a day.
It's almost like having yourown daily AI standup that's
been summarized in advance.
Kelsey Alpai (20:08):
Yeah, I love that.
A way to make the learninga part of your day, but even
just using AI to make thatprocess easier is great.
I love that.
Oliver Yarbrough (20:16):
Yeah.
And the beautiful thing isif you're already in Gmail,
a Gemini is built into it.
Mm-hmm.
So the, the beautiful thingabout having AI built into
something is you don't have touse automation anymore, really.
It's already there.
So it can summarizeit instantly.
Kelsey Alpaio (20:31):
That's great.
So we're just about out oftime here, but to wrap things
up for us, can you leave uswith one thing every listener
can do today to improvetheir AI skills and start to
future proof their PM career?
Oliver Yarbrough (20:44):
One thing that you can do to future proof
your career, I would say iseven though it's getting to be
old school now, it's been, it'sbeen out two or three years,
good old prompt engineering.
Okay.
And not just likeyour standard one.
Okay.
I would say that evenif you're a newbie.
You should probably, honestlyspeaking, start to learn custom
(21:06):
GPTs, even though it's a 201level, the basic level is gonna
be, it's gonna become so basic.
It's gonna be like talkingabout the internet like in
2025, like who talks abouthow to get online in 2025.
So I'm gonna encourageyou to advance from the
101 level to the 201 levelas quickly as possible.
(21:28):
Start learning customGPTs and more specifically
how to prompt for them.
When I say this, specificallyhow to write instructions.
Instructions, meaning thatyou're telling it you are
X, Y, Z, working in thissize company that becomes
basically your avatar.
You're basically, you'recreating a digital avatar
(21:49):
that you don't have to thenprompt it every single time.
That's the first level there,and then learning how to
write the prompt after that.
That's more of a workshopthere, but if you just
spend a little bit of time,take a couple of courses,
a couple of videos, learnthat it will do you wonders.
It will be a tool in yourtool belt that will give you
(22:10):
a superpower for the next 18months until everyone else
jumps on board and then itbecomes like a common knowledge.
Kelsey Alpaio (22:16):
Yeah.
I love that.
Oliver, thank you so much forspending time with me today.
This has been so helpful.
Oliver Yarbrough (22:21):
Thank you.
I appreciate it.
Kelsey Alpaio (22:28):
That's it for today's Retro.
Be sure to follow the showso you never miss an episode.
And if you wanna keep theconversation going with a crew
of 1,000+ project managementpros who get it, come join
us at thedpm.com/membership.
Thanks for listening!
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!