Ep 60: Unlocking Productivity With AI - ChatGPT 5, Automations, and Career Growth

Watch the YouTube video version above or listen to the podcast below!

Ep 60: Unlocking Productivity With AI - ChatGPT 5, Automations, and Career Growth Podcast and Video Transcript

[Disclaimer: This transcription was written by AI using a tool called Descript, and has not been edited for content.]

Dave Dougherty: All right. Welcome to the latest episode of Enterprising Minds. Got the whole crew here. Hope everybody's had a wonderful two weeks since the last episode. Like usual we start the the meeting and we start talking and then we go, oh crap. We should actually, we should hit record before we, we get too deep into this.

Alex has an update on something that he mentioned a few episodes ago, but honestly I forgot about it. So it could be a great refresher for things that. You might find yourself in, I know I've been thinking about this kind of stuff a lot lately. So Alex, why don't you queue us up with what you wanna talk about and we will do our thing.

Alex Pokorny: Sure. Also noting for those who are watching on YouTube or other video platforms, the black sweater crowd. I. Wait, we're so technically mine.

Dave Dougherty: Mine is the, oh, use a t-shirt on the court warmup. It's

Ruthi Corcoran: a dark, it's a darkly, it's a dark navy.

Dave Dougherty: All this is very us. This is a thing. No, you're wrong.

Alex Pokorny: Okay, I'll shut up. I'm sorry. No, you're good. All yeah, so the thing I mentioned probably a couple episodes back was talking about using chat GBT as a source for a 90 day plan for a new role. Actually we're at this company now, like six months. So it did an updated version of it of saying with my current role description and.

You know what, put together a 30, 60 90 and 180 checklist for myself, like a punch list of things I should know at this point. And had to rephrase it a few times until it got to the point of I need you to be asking me questions and then basically provide a space so I can answer it.

So that was became included in one of the prompt, and then I did a second version of it, which actually I haven't read yet, which was fill in what you know, for the answers. Of, and if you don't know, put in industry best practices and note where it's an industry best practice versus knowledge.

I haven't really told ChatGPT that much for context with my role. So what it sped out was surprisingly good though asked enough technical questions that I notes a little bit of the tool stack, so I was able to actually ask some of those questions and mentioned look over my LinkedIn.

That kind of stuff in the past. So it, it does know those details of my general role. I don't think my current role has much listed, but I know with my previous roles at least to be able to put together what a media manager, digital media kind of thing, role is. It was really good, to be honest.

It lined out everything on it was solid to be honest. All of it was a good, relevant question. A few of those things were. Yeah, I tried that it failed and yeah, I forgot about it. It never picked it up again or went down this path and. I haven't really been able to finish it up or, that's a new area that in general kind of learning kind of thing of learning new software, new platforms of, I know the bare basics.

I can log in, but I'm by no means could. Do the job at a company. If it was just asked to me to do it, I could figure it out. But I haven't put myself through those paces yet, so it'd be good to test myself that way too. So probably any place where you're at, it'd be an interesting test of saying someone who's 90 days in this job, or a first 90 day plan for someone in my role.

What's your first 90 days? It's a lot of meet and greets and kind of stuff that, yeah, I did the main ones and I skipped over the rest and then, may have talked about what my team does, but I didn't really hear about how their team could use my team or, at least rekindled that relationship to say Hey, how can we help each other

Using AI for Self-Reflection and Career Planning

Alex Pokorny: out?

Kind of partnership kind of stuff too,

Ruthi Corcoran: yeah. One of the things I really like about this, Alex, first, I very much hope that we share the two prompts that you used or revised and refined over time because they're, I'd love to play around with them. But what I was thinking about is with your, first 90 day plan for someone in your role.

If you are. If you're lucky or you find a a good team or a good hiring manager, you're gonna have something like a 90 day plan. Sometimes it's just the team's new, it's not feasible. Sure. And so this is a good substitute to just help you get focused. The other thing I was thinking about is, my role over the last year, two years has continuously changed, but without.

Without a formality to it. It is more subtle. It's oh, all of a sudden I find myself doing completely different tasks than I was two months ago, or even two weeks ago. Because of the nature of transformation in the digital world at this point. And so having a sort of help me through the first 90 days of this new role, I find my, in myself in and give me a series of questions to be able to think through how has my role changed?

What are those subtle things? That's such a cool tool for doing some self-reflection and helping orient when there's constant change.

Creating a Visual Resume

Dave Dougherty: Absolutely. I've, I've been meeting with a leader in my job fell into it but he has this idea of what he calls a Perkins menu or like a Waffle House menu or whatever you wanna, whatever your regional variant is.

Denny, right? For a diner menu that is mostly just pictures. And essentially what it's set up. Where you have five columns across your PowerPoint and you get like a little one sheeter, right? And those five columns are the five biggest skill sets that you have, right? And then you have three or four images representing whatever projects you've worked on before that align to that skillset.

That would be like the case studies that you would bring up in an interview or, just general conversations about yeah, I'm a strategic leader and here are the things that have, proven why I am a strategic leader. And it also prevents you from having just that huge chunk of text that so many people go into, right?

Where if you have an image and then a single bullet point that has to like only cover, half an inch on the screen you're naturally gonna be much more. Abbreviated with what it is. And essentially all it is that menu of options for when you're in an interview or a performance thing to be able to say, this is what I'm good at, here's why I'm good at it.

And, here's the example that backs up my experience in that. His challenge to me was to do that, right? So Alex similar to what you were saying with the, with your, reviews. I've also gone through and said, okay, chat, GPT, Gemini here's a sort of anonymous version of my resume.

Resume. Here's how I talk about myself and my experience. What am I missing? What am I underplaying? What am I overstating? What should I include in this concept of a Perkins menu? To move forward. And not only has that helped me refocus my own sort of personal narrative and how I'm presenting myself moving forward, but also to your point, Ruthie, like I, I was placed into a role a little more than a year ago that I wasn't super stoked about.

And but I needed to track. Okay. Mostly just for me 'cause nobody was asking for it. But for me and the way that I like to work and challenge myself, I needed to know that I was making an impact, right? So when it came up to that year point, it was, alright, is Dave worth the time? Yes or no?

Let's put those things together and. And do that. A similar thing and related thing I should say using.

Ruthi Corcoran: And the thing I was thinking about with that one too is similar to the 90 day, going through that exercise periodically, whether or not you've had a specific role change or an informal one or things are just shifting around.

That's such a. It's a tangible way in which you can say, okay, here's what I'm good at, or Here's where I think I wanna focus. What have I done in those areas? Or you could reverse it and go, Hey, here's the things that I've done recently. How do we think about those in a new way? Because as we evolve in our careers we have to shift our intuitions about what we think not our, not just shift our intuitions about what we think we're good at, but also there's. Dare I say, an identity change around there, like I think I'm this, but, oh shoot, the last four projects I've worked on have actually shown me that I'm this other thing, and that might all be a better jumping off point to where I wanna go next.

That's so cool. I love these two.

Dave Dougherty: Alex, you're thinking I can see that you're thinking, yeah. Yeah. You're terrible at

Alex Pokorny: poking. Killing me. No, I was just thinking about this in a future forward statement and one Dave, for the kind of the resume review kind of piece, especially for those out there who are looking at jobs.

You can also ask it since it has a rather large wealth of knowledge of saying, look at similar job descriptions that have been posted, and then what are key bullet points that I'm missing or under emphasizing, something like that because, or what language do they use that's different than mine?

Because you might be using in-house knowledge and phrasing, right? Versus industry standard or new phrasing. We've talked before about how it used to be called e-marketing, then it became internet marketing, online marketing, digital marketing. It, it changes all the time, like

Ruthi Corcoran: and marketing, calling it.

Alex Pokorny: Oh my gosh.

There will be a day sometime where someone says the similar phrase of saying, instead of the facebook.com, drop the keep the facebook.com. Someone else will be like, ai, that's too long. Drop the, yeah, that'd be something else. It's just

Dave Dougherty: intelligence now. 'cause we've outsourced it all.

Yeah, it's not artificial.

Ruthi Corcoran: It's all

Dave Dougherty: hybrid. No. Yeah. Yeah. An optional extra like next step that I had fun playing around with along these lines was Ruthie, to your point, the here the things that these projects have pushed me toward.

The question for me was, okay, I've been in these roles.

If I want to be known for something else, how can I resear myself toward the thing that I want to be known for? Or have people, recognize me as okay, Dave's our guy for that particular topic. And less so this other thing. I went into chat, GBT and I asked and Gemini I do my testing in both environments just to see who does what and who does things better.

But I asked them, okay, for these next roles that we've talked about that I might be interested in, create a list of sort of day-to-day. Activities related to those roles, and then breakout of those day-to-day, what can be replaced by ai? What can be AI assisted and what is purely human? Because then I want to double down on the purely human and lean into the AI assisted stuff, right?

Because if I'm trying to pitch myself as something that is. Fully going to be automated very quickly. Then, that's just a losing formula, right? So then how can I rep those things slightly differently, really.

Ruthi Corcoran: I really like this approach and I continuously appreciate your comments about, I wanna focus on what's human and I wanna double down on, on those elements.

Balancing Work and Life with Automation

Ruthi Corcoran: The project that I've been embarking on in my minimal spare time in the last month or so has almost been the reverse of how much of what I do right now. Can I automate? Yeah. I want, I, and my motivation for this is I am spending more time. On work than I, than what I want, right? I wanna be spending more time on life things.

I tend to over, huh? I'm not quite in workaholic status, but I'm, if it were a scale, I'm gonna be closer to the workaholic than somebody who just sits idle. And so one of the things I've been thinking about is, hey, if I want to clock out at three every day so I can pick up my kids and actually spend time with them during the waking hours when we're all coherent and not crabby.

How can I maintain a similar level of output with that sort of truncated timeframe? And so I've been pushing myself to, okay, how can I create an agent that can assist these tasks? What are the things I can just completely, like, how can I hand off the things that I'm doing? Where am I spending too much time?

Maybe I'm overthinking things. Maybe I'm just overthinking is usually the situation and how can I hand that off or just get better at, let's just do, let's just do good enough. Where it is less important.

Dave Dougherty: That's interesting and I'm glad that you're trying that. 'cause Yeah, we all need balance.

I think the one thing for me, when I got over feeling the need to respond immediately. I know for a lot of people that can be a huge endeavor to get over that that worked wonders for my stress level. So like I check my email maybe two or three times a day throughout the day and that's it. And then that forces me to prioritize what has come in.

And what doesn't, and I usually check it right before lunch and I prioritize lunch over everything. If you're backing up against lunch, then you know, only the important stuff gets done. I'm just, I'm talking what works for me. You find your own thing. But I remember talking to some executives recently and they were all talking about what granola bars they keep in their desk because they they skip lunch.

And that is such an anathema to me. No, I just, I'm not gonna do that. And if you won't need a meeting with me over lunch, guess what? We're gonna be eating when we have that meeting.

And then like team stuff. I do a little more with teams things, but I found as I got into like global roles, I was getting pinged so frequently that the notification bell became a trigger. And I know certain people who have been on Slack channels, for their companies, have the same problem.

So I turned off that notification too, and it's just been awesome. Now, does that mean I'm not doing my job? No, that does not mean I'm not doing my job. I am completely doing my job and I have very frequent check-ins to the people that matter. As to whether or not they think I'm doing my job the way that I should be.

And the answer is always yes. So for those of you who are feeling super agitated and nervous right now, after hearing me explain this I'm glad you've recognized that feeling in yourself. Now go do something about it. And that's probably a sign that you should probably execute some of this stuff and just chill a little and go get lunch.

Go get lunch.

Ruthi Corcoran: Lunch. I've stopped taking lunch since not being on your team or wi working with you? I just the number of lunches I take has plummeted.

Dave Dougherty: Let's fix that. Let's fix that. Yeah. Yeah.

Ruthi Corcoran: Alex, any thoughts on that stream? I've also a new one too.

Alex Pokorny: Yeah. Tough with a new role.

Got that imposter syndrome pretty hard. So it's always trying not to ever mess up at the same time. Like, how do you know to know where your baseline is? I realize I'm always tough on myself too, so it's always I realize I have an underlying bias. But there's always a point too of how do I kind of show.

Progress, especially in an organization of constant change, and it's pretty difficult to actually get something done when the organization keeps radically shifting. Basically even within six months, it's. Like hardly the same some of those notifications and stuff is like the only way, but it's really funny because my direct report, and I were talking about this recently of email doesn't really seem to work.

Like the number of emails that just go unanswered is pretty frequent. So instead it's like you always have to set up a meeting which nobody likes their calendar being full of meetings, but at the same time. Or else are you get their attention or Teams Chat has been like the best way to get things done, but it's also the most annoying.

It's like it it's also easy to get into like casual conversations and like just keep on going and stuff like that too, of trying to get back and refocus back to work where at least the meeting ends at a certain point.

Dave Dougherty: That is one of the benefits of working in person again, is that I can.

Granted, I, I look the way I do so that, that helps. But going into somebody's cube and just saying, Hey, I have a question for you. And if they ignore you, just standing there,

say, Hey, you wanna get rid of me? Answer my question.

Alex Pokorny: Yeah. I think that's a struggle. I don't know. I've heard it from enough people and I definitely feel it myself too, of, when is it a meeting, when is it an email? When is it a chat? When is it like, all these other things and then the misuse or believed, perceived misuse of one medium for the other.

I don't know. I think everybody's all over the board and it's kinda gets down to company culture, but also people's own expectations and experiences too.

Ruthi Corcoran: This is exactly what I'm talking about, guys. Outsource all that little nonsense.

Alex Pokorny: Yeah. Just

Ruthi Corcoran: outsource it. Do I make it a meeting? Do I make it a thing?

I don't know. As che,

Alex Pokorny: G-P-T-I-I gotta get more into. Using voice to write my emails, because number times I've had an email basically written in my head over and over again, you, I finally actually write it, and then I finally actually send it. Those last period in the last two chunks are a little bit too large.

It's like I already know what I need to say to the person. It's three things. It's oh no, I'm gonna spend forever writing and crafting this email. That was a big unlock

Dave Dougherty: for me. Actually, I had a previous. Manager who's okay, it doesn't matter if you got it done, if nobody knows that you got it done.

I said, okay. That's fair. Especially if you wanna move up in the organization, right? You have to be, yeah. You have to market what it is you've accomplished and make sure that everybody knows it, right?

But yeah, doing the voice to text in Word and then copying, pasting that over into Outlook was a complete game changer for me because it wa it, it literally was the act of typing was preventing me

Ruthi Corcoran: from

Dave Dougherty: sending all of the emails I probably should have.

And yeah. It's like I have all these microphones in my studio. I, it just makes sense that I can talk a little more for it along these lines too.

Exploring Automation with Make.com

Dave Dougherty: Of automating things I've been trying, like last weekend before, I've been trying make.com. I know we talked about this a little bit a number of episodes ago with N eight N or Make or some other things where you can finally connect things together.

And I gotta say after. Three days of playing with it. The promise is there, but there are so many sort of optional things that in order for the automation to work properly, you need another subscription or you need access to the API, which is a higher level of subscription than you have, or, all these different things where you're just like, oh.

Okay, so I know what, I'm smart enough to know how to do this and that this is a thing, but I don't have the capability to just spend my way to the solution, and of course, halfway through the three day timeframe of trying everything out was like, is this actually better than me just doing it?

Because this feels a lot longer, but it was interesting. I, I don't know if you can see the whiteboard behind me, but I did have things mapped out and was just, I did have a realization that a Google sheet was capturing a lot of the information and was the central hub for things.

I'm like, okay, if I tweak things a little bit, then I could leverage that to drive all of these other things. But part of the three days was standardizing the information too, right? That's one of those problems with the automation is having all of these things standardized so that the robots know exactly what you meant and there's no wiggle room and there's no, this is the way that, Dave does it.

It's no, this is the way the robots need it to be so that it just works. And am I happy I tried it? Yes. Is it done? No. Will I keep going? Maybe. I think even if I could automate one or two things, even just to up automatically update the table, that could be really cool and that would save me a bunch of copy paste time.

But I'm also not about to spend $200 a month for enterprise GPT. No, just not. I'm just not. I know Alex, you've been trying to do some automations and I know you talked about your experience with this before. Are you still doing that? Have you tried anything new recently?

Alex Pokorny: Yeah.

Alex's Experience with Looker and ChatGPT

Alex Pokorny: Actually, the project I've been working on is this Looker report that pulls in a bunch of different data sources. Been using Chad GBT five, which is actually pretty funny because it's just got released and it basically auto selects which model best fits your question. Yeah, some critics or detractors are saying that it's just a cost saving measure for.

Open AI because they're pulling different models in, but also the model selection. There was really no naming system before that kind of thing too. So I think there's enough confusion there in the market that. Pretty valid too. It's not an agent though, and I was playing around with an agent before it's really funny.

So I have three chats who seem to have no recollection of each other even though it's shared memory. One of 'em is five, one's four, oh. One of 'em has browser access, but not agent access. The other one is an agent, but it doesn't seem to have browser access. And then I have five, which apparently it has neither.

Ended up like manually, step by step, creating it myself with luckily with chat, GBT tutoring me and teaching me along the way of as to teach me like a beginner, give me all the step-by-step of exactly where right buttons are on the screen. Give me everything because.

Screenshots would be great. Screen share would be great, or something like that, but I don't have that option. So at least descriptive text would be good too to try to build this out and been doing it. But man, it's slow. And I was gonna bring this up today, but I. There's a funny Ethan Molik is a professor blogs blog about ai.

The Beautiful Mess of Modern Business

Alex Pokorny: I mentioned him before, but he has a blog recently that was called something like Around a Beautiful Mess. And it's the model is a business model, which is called The Beautiful Mess. And it came from this idea of, or this experiment that I. College did with the business of mapping out exactly how they built their thing, how they actually manufactured their item, and they decided to map out everything.

So they, the photos of it, you say you got a whiteboard, you should see the random strings and all the rest kind things. Background the photo alone is beautiful. But it's crazy. Basically there were completely redundant processes. There were shortcuts that were being used actually.

In the, how it's being built. It's not a shortcut anymore. It just literally has shifted over to that. There, there was tons of things like that and crazy loops and pointless processes and everything else. And then in the end it made the item, right? So there was a project manager brought the CEO through it and to use cleaner language.

He was like, it's even more messed up than I thought. It was like nobody knew how bad it was and became this. Business model I'm talking about basically is modern business is super messy. It's astoundingly messy, like the process. And if you actually were to actually ever write it down from beginning to end for any project, anything launched, you'd probably be like, why?

Do all this.

AI Tools and Their Real-World Limitations

Alex Pokorny: And I think that's I've been playing around with some AI tools and that I've struggled with some that can't do a website rebuild. They can do a website build and they give you very much what I would call like a Squarespace build of like chiropractor and San Diego. And they give you a testimonials page, like five pages standard.

I'm, if I'm trying to re-skin like a corporate website that exists in a different tone, better usability, WCAG, accessibility standards, like all these other kind of things being applied to it. That's too far because it's too messy. That's just so much. It, it just was so fitting to what I've seen as well of, there's so much promise with a lot of these AI tools of what can exist in a really clean environment.

But the moment you throw in like a real world environment, it's I can't do that. 'cause I don't have admin access of that. I do have access here and I can use this tool and I can be scrappy. And it's like everything becomes these weird shortcuts and strings through it. I don't know, tools like make I think are really cool.

I was playing around with Zapier and make for a while and it was a similar situation where basically it was like, okay, now I can connect these data pieces together, or there's a bunch of blocks in between from me actually ever doing this particular project. And then, tech issue after tech issue and that kind of stuff.

Ruthi Corcoran: Yeah.

Alex Pokorny: Weirdly that it runs into job security. To be perfectly frank, someone's gonna be doing this, but goodness, my boss isn't gonna be doing it. It's gonna be me.

Dave Dougherty: I was surprised when I was doing the make.com like I wanted,

I wanted to be able to take the text that's created by one thing into a transcript, and then have that text extracted and put into a sheet. Sure, sure. Sounds basic enough. According to the make.com stuff, maybe I'm open to the fact that I might be an idiot on this and that's fine. The only thing that could possibly do it was Google's vision cloud like plugin to be able to extract and format things to then send off into other, and I'm like, that's strange, right?

And also when I went to chat GPT and I was like, can you transcribe an audio file for me? And I was like, yeah, absolutely. I can do it. And I was like, okay. I upload the audio file and it's actually I need to be attached to Whisper and I need this and I need that. And I'm like, so you can't, so you can't, I don't need a, technically I could, I need a yes no.

So that's been the really interesting thing where it's okay, the tool stack that I have works, and even though I'm playing with it and considering changing certain things in the next year, it just works. We have it down. So is it worth it to learn the new system at this point?

Nah maybe not.

Ruthi Corcoran: This is so great. I am in such a different perspective than both of you.

First, I have to give a shout out to. Tim Hartford's book, messy Creative Resilience. In a Tidy minded world. Fantastic book. I've, I think I've reread it twice. And if you're, if you like books where each sort of chapter is a self-contained article that explores a different topic, this is one of those which I very much appreciate.

'cause then you can just pick it up every once in a while. What he talks through are all a number of case studies across music, across the tech world, across MITA whole, a whole range of different case studies in the modern world in which. It's the messiness that allows for or opens up the space for creativity.

And how that messiness might not be. It might be an important part of the creative process. Maybe it's important for that product to be made, like you can't get around the messiness. So highly recommend that book. Just if you need a different way of thinking about the messiness of our everyday corporate world.

The other thing that's coming to mind is I haven't tried make.com so full disclosure, and also I think I'm. Probably a month or maybe two months behind both of you in terms of what I'm trying to do with all of my AI use cases, and I think in this case from a mental health standpoint, that might be healthy.

Because I, I might have tried to do something of. Few months ago and gone, oh, it's not quite doing it. And then I just let it go and now I'm trying it again and I'm going, look, it does the thing I tried to do a couple months ago that I couldn't get it to do, and now it's doing it.

This is amazing. I have a few extra bags around my eyes this morning because I was up super late because GBT five launched. But not only did GPT five launch for open ai, GPT five launched with copilot simultaneously and the improvement to copilot. Has been substantial just overnight. And that's been really cool to see.

And specifically some of the things that I've enjoyed playing around with is, hey, I've had a couple of agents I've created for myself just doing basic things, i'm not trying to have them execute things. I'm trying to have them help me create various assets or excuse me, artifacts.

Or help me think through or help me plan or just Hey, I can't remember what I did last week. Can you please just give me a recap? And turns out it's getting much, it got much, much better at doing those types of things. And it, the messiness is a. Cool tie in because one of the things I observed is it got overnight, much better at parsing through the messiness that is chats, emails, SharePoints documents, et cetera, et cetera.

And that has just been so cool to see. And even just simple things like helping me create A-J-S-O-N instruction set to give my copilot agents so that they become more productive. And perhaps even being more thoughtful about what it is I'm trying to do with the agent than I was which is both scary and exciting at the same time.

That all has shifted just within the last day. I'm curious your thoughts on GBT five, knowing that you've reached some frustrations with its age agent capabilities. What are some of the other initial takeaways since it launched yesterday?

Dave Dougherty: Dave, do you have any or do you want me to run? Yeah, I'll just, I'll state for the record I've been in a basketball camp all week.

So the farthest I got was reading through Ethan Mulnicks or Malik's newsletter about his experience with GPT five and how

Ruthi Corcoran: it just does stuff.

Dave Dougherty: Yeah, it just does stuff and how it, like it wants to help you with things, which is great. But having a golden retriever, I have enough of that in my life already, where it's just alright, stop trying to be helpful.

I love my dog, but seriously. And in that newsletter, the thing that jumped out to me and I think Alex, what you mentioned it maybe in a slightly different way, but it seemed that it was selecting whatever models fairly arbitrarily. And the main reason for that was it has to decide how difficult.

What it is you're asking it to do is, so I'm very curious to play around with that and to see that kind of, play out through things. But yeah, that, that's that's where I'm at. I haven't had a chance to sit down with it fully. Alex what have you

Alex Pokorny: done? Yeah, it definitely needs refinement. I would put it out there. It's a seen improvement above four, but it definitely needs some better refinement. Any, if you played with four four oh or any kind of variation on those they all kinda had their own. Limit, upper limit. And it was it the more technical or the more detailed a response that you wanted, the more evident that was.

If you looked at 3.5 versus four, though, that was a huge step change. And I think four to five has a step change as well. So there, there's still improvements in what's being done there. And the project I was working on, like the descriptions that I was giving was far more accurate.

It was. Far more complete. I've been actually testing a lot of the free version versus four four oh. Just literally yesterday and the day before, figuring out their knowledge base around particular topics. So I was testing them with all these different queries and incognitive windows and VPNs and all the rest of the kind of stuff of trying to see can I update its core based knowledge, which is actually still kinda a limitation of.

These systems of how much they can learn as well. So to answer real quick, I think it's better if it's picking the right model every time. I don't think so. I think it's underestimating, or even sometimes, to be honest, overestimating some of these questions. Like it's taking way longer than it should for some stuff.

Other ones it's still struggling with, but four would completely fail on it. So it's able to complete things that it. Couldn't do before, still a chop change. Still more to go, but definitely better. Let's be honest, also, Ford didn't release that long ago, so I just gotta wait a couple months.

Dave Dougherty: Yeah, it is crazy. Ruthie, to your point earlier on, just how fast everything is changing and even just trying to stay up to date there is a burnout point. Yeah. And I was thinking about that actually while I was watching my son at this camp because, you can't download an app and learn how to do a crossover dribble, right?

You have to just learn how to do it and put in the reps and just do it right. I know the thing, one thing that's jumped out at me with your 30 61, 80 comments before Alex, I think was you were asking for a 360 review and found out your coworker was using GPT for it. Oh yeah.

Yeah. And that.

That to me, that has stuck with me and I've been trying to figure out why that's bothered me so much. Yeah, I should probably just write a newsletter on it to get some thoughts down on paper, but it just seems like a breach of trust, and on the one hand, yeah. Okay, cool. We can get all these answers, but what are you gonna do about it?

It's one thing to get all the answers and to put it out, but if you don't actually know how to use it or if you don't know how to actually apply it, then we haven't actually improved anything. We haven't moved forward in any way. And I was thinking about that with my kid. 'cause there's, loves basketball, but there's a 0% chance that he gets any taller than six three.

I will put good money on that statement. So I go to the coach who has coached at a very high level, WNBA and European basketball and, like serious levels. And I just said, Hey, for the shorter dudes, what what sort of skillset should we like, focus on?

Or, encourage, more than the other ones. 'cause you're never gonna be a rebounding champion if you're, two feet shorter than everybody else. But that led me down this rabbit hole of going to Gemini and doing a deep research to say, give me the top 20 short basketball players.

What made them different and how could I incorporate that into a practice routine? That was wonderful and I did that in 10 minutes and now I can read through it and figure out how to incorporate that into, my son's practice routine. And before any of you email us saying, Dave's forcing his kid to do stuff, this is.

You don't know my kid, so just know that he's driving this, for sure. This,

Ruthi Corcoran: so

Dave Dougherty: yeah.

Ruthi Corcoran: This reminds me of a conversation we had, maybe it was a few months back about it is not going to give you the questions. It will give you the answers, but you have to know the question. You have to know where you want to go, and that I think, is continuously reinforced with everything I've been working on, the cases in which I have a clear idea of here's what I'm trying to get out within the parameters of what the technology can do.

Of course not, as I mentioned, I don't push the boundaries too much in the cases where I have a clear idea of here's what I wanna do. It's a very powerful tool. In the cases where I'm floundering, it can also be a powerful tool if I'm asking questions of, help me how to think through this problem and how I should be utilizing tech, these set of technologies or how I can even.

Reframe my thinking. It's, it is powerful in that case, but ultimately I know where I wanna go. I wanna change the way I'm thinking or I wanna get to X point. And I think your example of the 10 minute basketball case just exemplifies that point.

The Importance of Knowing Your Goals

Dave Dougherty: Yeah, because, and I mean we've talked about it a lot on the show and I'm sure we'll come back to it again, but I think these tools are only as good as where, you want to go.

If you don't have the goal setting. If you don't have an idea, then they'll definitely help you out. Try to, formulate an opinion to figure out where you might want to explore. But you still have to work. It's exactly what I said to the students at the university when they're like, networking's hard.

Any way that you know to make it easier. No, it sucks. Go do it. That's it. That's the advice. Just get used to it because nobody else likes it either. And that can become your superpower, to be the one that sets up the coffees.

Ruthi Corcoran: That's one more point on this. This you need to know where you wanna go and then there are very powerful tools.

Fun with AI: Pigeon Simulation

Ruthi Corcoran: I do have to share that part of playing around in the last 24 hours with GPT five has been the semi fulfillment of a long term. Dream goal of mine, which is to create an application. Originally I envisioned this to be something you use in sort of a VR scenario in which you're in St. Marks Square and you have the opportunity to kick the pigeons that are polluting Venice and all of its beautiful sculptures.

I didn't ask Chi Chat, GPT to be so cruel. Instead, I said, I'd like to see a simulation in which you can click and feed popcorn to the pigeons and the fact that. Within less than five minutes, I could ask it to do that run code and see my little pigeons moving around. The screen was phenomenal like that.

Just that made my evening.

Dave Dougherty: I'd just like to state for the record, Alex and I do not promote violence against animals.

Oh, leave the flying routes alone. Come on now. Exactly. Yeah. Or as my niece calls them, city chickens.

Ruthi Corcoran: Ooh, I like that one.

Dave Dougherty: All right, on that note with PETA calling we will, we'll see all of you in two weeks. On the next episode, please go subscribe to the Pathways newsletter. We are adding great stuff and we added over a hundred subscribers in the first week. So thank you to everybody who did that.

And we will try to get Alex to share his prompts for you to play around and we will put those in the Pathways newsletter. So go get those, sign up for that, and join us in two weeks for the next episode. Take care.

Alex Pokorny: Think so.

Ruthi Corcoran: Cheers.