Ep 75: Claude, Cowork, and the New AI Oh Crap Moment
Episode Summary
In this episode of Enterprising Minds, Dave, Alex, and Ruthi unpack their latest experiences with Claude, Claude Code, and AI-powered workflows. Dave shares his first true “oh crap” AI moment in years after using Claude to turn a rough business idea into deep research, strategic analysis, and polished deliverables in a fraction of the time traditional research would have taken.
The conversation explores where AI is genuinely useful right now: speeding up research, supporting project management, handling repetitive analysis, and helping people move faster on ideas that would otherwise sit untouched. Alex frames AI as a tool that makes certain tasks easier, possible, and scalable, while raising a bigger question: what parts of your work would you be happy to automate?
Ruthi brings the counterbalance, pushing on whether faster output actually leads to better decisions and how much of this promise is limited by enterprise tools, culture, and change management. The result is a grounded conversation about AI’s real value, its current limits, and how knowledge work may change as these tools become more integrated into daily life.
Ep 75: Claude, Cowork, and the New AI Oh Crap Moment Podcast and Video Transcript
Ep 75: Claude, Cowork, and the New AI Oh Crap Moment
Claude Oh Crap Moment
Ruthi Corcoran: Ooh.
Dave Dougherty: Hello and welcome to the latest episode of Enterprising Minds. Got the whole crew here. And today we're going to discuss Claude and our recent experiences with that. Honestly, guys, I gotta say that with Claude, I've had my first oh crap moment since chat. GPT like hit. Hit the market. Most of it is just playing around with cowork and we can get into that.
But
Alex Pokorny: Oh crap moment. As in man, the bots are coming from my job a thing. Or
Dave Dougherty: yeah,
Alex Pokorny: really cool. Or what.
Dave Dougherty: it's the, this is cool, but also the, oh crap, the implications of this, but, and this, we'll get into the conversation with this as well, but it's for the, luckily having gone through that with chat GPT and being what, four years now into this, three years now into this I'm realizing, yeah, three I'm realizing, okay, that is an initial reaction, right?
And. I am constantly surprised at how few people are still leveraging some of the basic uses of ai. So I know that my initial reaction is not necessarily how it's going to play out, but for anybody who is taking advantage of all these things. Wow. Now. I'll also caveat everything that we're about to discuss with Alex's experience in the previous episode with some, oh, hey, this isn't as good as everybody's saying.
So if you want to do a point, counterpoint, go listen to the last episode.
but yeah, alex, Peru I'll kick over to you unless Ruthi, you feel strongly about something and want to jump in straight away because
Ruthi Corcoran: I want you to walk through what was the experience you had with Claude, and then what was your reaction to it, because about what you've encountered.
Dave Dougherty: Alex, you want to go first?
Alex Pokorny: go for it. Dave, what was your Oh shit moment.
Testing A New Business Idea
Dave Dougherty: Okay, so for me, what I've been doing is I've been teasing out this. Business idea. And it's something where on the surface of it, I don't have a lot of experience in the thing. And even when I tell my own wife, she's what you want to do? What? Why? But I feel strongly about it. So I'm testing the idea, I'm fees, feeling things out.
And I've been a big proponent of Google Deep Research and the ability to do Google Deep research for a long time now, and it's been unbelievable. So I started the usual way. Doing Google Deep research stuff and then putting it in a notebook, LM and then getting, some additional resources there just to get like directionally correct.
I'm not worried about is this a hundred percent factually perfect, but will this send me down a particular path to become more knowledgeable? That's what I was looking for, at least with that. And it was useful, and like right around that time, Gemini or. Put in the deep research prompting into Notebook lm, which is nice.
'cause then you don't have to sit and then port everything over. It just puts in all of the sources from that deep research into that notebook so that you can play around with all the sources it grabbed from. So that's pretty cool. So then, we were hearing all this stuff about Claude and Claude Code and then Claude Cowork, and I was like, okay, I've been feeling like Claude has been giving me better outputs, at least for what I've been using it for.
So I went ahead and just bought a subscription to it. So I now have the paid subscriptions for the big three.
Cowork Connectors and Skills
Dave Dougherty: And then with Cowork, I just gave it some basic stuff. Okay, here's why I've been using chat GPT for, I'd like to recreate these custom GPTs in, projects or skill sets or, 'cause it's a slightly different.
Set up to GPT or Gemini and everything's got its own feel, right? So as I learned more about the way that it's set up and what you can do the more I'm like, wow, okay, cool. This is not necessarily a one-to-one, but it's a step up, in addition to being able to execute everything that I said there, or wanted to have it do additionally.
I went and looked at all of the different connectors that you could do with it. So all of these different apps that you could put, tie into these things. So there's mural and HubSpot and all the Google stuff, and then Word and Excel and then, similarWeb and SEMrush and I remember I texted you guys this, when it first hit me where I'm like, holy cow.
Just with the ability to do the custom, skillsets or projects or whatever, plus these connectors just with that's. A small to medium sized businesses marketing department. Asana was another one that connected into it. Then I started learning more about skills within Claude and how skills work in their platform versus a custom project with a custom instruction or whatever.
So then I started clicking into all the ones that Anthropic made, because I don't want some schmos code that I don't know who they are, at least Anthropic is an entity that is Sueable. So I clicked through and added through the ones that were going to help me for how I primarily use Claude, which is project management, marketing business strategy, marketing, like all of these additional skill sets that you can get within the tool.
Which then made things even more specific and specified because now whenever I say, Hey, help me do this, that, or the other, it's now going and looking at the custom instructions, looking at all of the tools that I've connected, looking at all the skills I've put in, running through all the skills, and then where I've done.
Any of the custom ones are uploaded, brand standards or uploaded whatever else it is. Now going from idea to research to branded output in 10 minutes, 20 minutes.
22-Minute Deep Research Demo
Dave Dougherty: So the other day I had a very specific thing that I wanted it to look at and I said, okay. Now, mind you, this is while I have a cheesy James Bond movie on.
So it was like, I'm watching, I'm not, this is my downtime, but I'm, playing with an idea and, I put in the this isn't even a cowork, this is just normal. Claude Chats, but tied into a project, I had the research button put on it. I had the web search thing enabled and I just said, I want you to look at company A, B, C, D.
Go find, go through their investors section of their website, look at the annual reports and any of the SEC filings that they've done. Tell me about their journey from when they first got. Seed funding to where they are now. What worked? What were the bad decisions? How did that play to that In my situation and my idea, what can I learn from that?
What should I avoid? What might I model? My way of doing things around what has previously worked and then also unrelated to those public ones. Go look at these, the, this private one to then look at this other idea that's related, but could be another direction of doing some stuff.
And it went through and in 22 minutes it looked at over a thousand documents and websites. It asked me how I wanted it formatted do you want it formatted previously with the footnotes and the citations and, dah. And I said, yeah, I want it the way that we've talked about it. And, I've even started like a do not ever cite this, this resource because I don't trust any of it, so don't do it.
Within the fight scene of goldeneye, I went from idea to a Word document version that was 23 pages, three sections with tables and visuals, and a PowerPoint version. That was 15 or 20 slides, which was a more visual version of the Word document. I just went, oh my God just, if I was to look at the websites it went through, that's six months because of the day job and being a parent and being a friend to people and being the kind of human being I want to be in the one life I have.
So that is an insane amount of work that I can go test stuff and go look at, and I'm just like. This is not even getting to the point of go execute stuff like with what I'm doing on the podcast and repurposing content and doing other stuff. It was just like, oh my God, this is insane. Because the time to launch something now is going to be crazy short.
So anyway, immediate thoughts, comments, questions? I've been talking for a while.
Ruthi, you look the most uncomfortable. Why don't you jump in?
Ruthi Corcoran: Oh, Alex actively pushing buttons and then took a sip. Okay.
Ruthi Value and Limits
Ruthi Corcoran: So immediate reactions are what you just described. I think your time to launch comment is good. It's super important that you know what you're going for,
Dave Dougherty: Yeah.
Ruthi Corcoran: because you could do all the research and all the synthesizing in the world. to what end and where is it adding value? So that's my initial reaction. So it goes, okay. So I've seen a lot of that sort of research activity corp in the corporate world in which you're spending a lot of time sort of creating a report, doing an analysis be that in the competitive landscape, or you're trying to understand the technology landscape or, insert landscape here. And so that cuts down a just a ton of. frankly. Sometimes it's fun tedium to be going and exploring and so that sort of cuts it down, synthesizes it, assuming of course that, you've set it up in a way that can double check like the outputs and you can trust, trust basically what it's saying. So all of that is super useful. And I'm still left with this question of but what are we doing with it? So like I can think of a number of applications within the corporate world in which you go, okay, now I can much more quickly maybe answer a strategic question that I'm trying to understand so that I can then make a decision,
which is super important.
That's a huge value add. Where I think I have less of an oh shit moment though, is I go. Okay, we like you made that faster, but it's not, I don't have the, oh my goodness. Response this is actively something that's going to replace or otherwise shift parts of the landscape. I guess that's the part where I don't see the connection as much of.
It's doing the super powerful thing. I can see it being very valuable in particular instances and expanding the capabilities that I already have, but I don't see it as replacing just yet. Where it does get interesting. So that all said, I guess I'll counterpoint myself, is if through that synthesis you're able to see opportunities that you didn't see before or that you quite literally couldn't see before.
I know in the biomedical space, for example. AI is being used to just do a ton more scaling of research in terms of what are potential viable molecules that we could be applying, right? Or how do we think about clinical trials and improving the efficacy of the next clinical trial we're doing.
So they're changing the landscape in that way. And that's pretty phenomenal. So
that's my initial reaction.
Claude Code Printing Press
Alex Pokorny: Lemme throw out one thing. There's, worth listening to, if you're interested in diving a little deeper Lenny's podcast. He has a recent episode with the head of Claude Code. The guy who developed it, actually it wasn't very popular about product internally a long time because it was going in a different direction than they wanted, most people were thinking that it took off.
And then Claude Code, if you know the history of Cowork on Code built Cowork
in a couple of weeks. So it was code. Cloud code basically being re-skinned basically. So it was being used for non-technical reasons, which there's a whole list out there that Lenny has as well. But this guy Boris Cherney, really interesting guy to listen to because he also was trying to pull a lot of historical parallels and it got to this thought experiment, which I've been turning over in my mind the last few days.
And I still don't have an answer to it, but it fits right where we're talking right now. So I'll bring it up of. He was trying to come up with a historical parallel to cloud code, basically taking over coating, he's seeing programming as salt, just salt. The number of commits now to GitHub that are actually straight from cloud code is large and rapidly growing, so it. Taking this stuff on to the point where he gets people from X or Twitter stating a bug, he'll spin up. Which it was interesting also. They've totally verb Claude by the way. So he'll spin up a Claude and they're clotting on things. He'll spin up Alaw to work on that particular bug while he goes, reads the next comment, and pushed out bug fixes under five minutes. Which is astounding to me because that could totally go to a customer support role If you could find someone who could actually do that's a totally different mentality. the analogy he was trying to use to explain this new world was, the printing press and the changes it made there where. The ability of literacy was, small percentage of the total population, and that was the Dark Ages Project Breast gets invented. It's one of the main reasons that we went to the Renaissance. suddenly books are everywhere. They're a lot cheaper. The common populace is able to read, and ideas are exchanged, letters are exchanged.
Suddenly things changed very drastically. And one of the comments that he was relaying was from some of the scribes who used to write the books. Were very happy for the printing press because the part that they hated writing it. The part they loved was illustrating the books,
so they got, they were able to get rid of the part that they didn't like. And for him, it's a very similar thing. He's through and through. He loves it. Like he goes on vacation with his wife and he spends most of the time programming. And that's his favorite vacation's. That's does one of his favorite books is all about explaining a code language that's not used as frequently anymore. Loves the stuff. So Lenny asked him, I was like, you, when was the last time you actually wrote any? And it's been months.
so how does that make you feel? And his reply was that was the part I didn't really like. creating stuff I didn't like. That the right semi was in the right spot and the right line was in the right carrot was in the right spot. I didn't like that. So this actually allowed me to get rid of the part that I didn't like and that, that's the thought experiment.
Automate The Parts You Hate
Alex Pokorny: Now that I've been turning over my head the last couple of days is okay, CLA can do analysis. All I can do, put together ad copy, it can review and do some UX and conversion optimization some of that stuff is, personal critique and debatable, but pretty solid work. So what kind of parts of my job do I not like? And those then are probably worth finding a way to automate, like you spent instead of six months on a project or research project, which might have been fun for
The time cost was too high, claude made sense there to take on the time cost to take away that thing, to give you the output that you wanted because it allowed you to move at the speed that you wanted to move. So that's kind of the other element of it, is like I think of AI now, as it will make certain tasks easy. It will make certain tasks possible; it'll make certain tasks scalable. It doesn't mean the output's going to be good all the time. That's not, quality is not a part of that. It's a thing that it makes it easy. Possible and scalable. Recent, recently I had a list of 50 keywords and I said, what's the best URL on this website for these 50 keywords?
No crawl data, no nothing. Go out and go find it yourself. And I was able to do in first column of the best URL it thinks for the landing page for all of these 50 keywords somewhere. Type it on Google. Here's the best one. I it faster than I could and I was doing something else while doing it. I was very happy for it to manually run through the website and find stuff.
Dave Dougherty: Right.
Alex Pokorny: Had some edits afterwards, but it was great. So, I was happy to let go of that piece. And that's the piece I keep now considering is what other parts what's the parts I like? Maybe it's the output, maybe it's the process. Maybe it's the people, it's the interaction. There's always, there's other pieces too, but is there certain parts of my job that I'd be very happy to give up if I could?
Prompting And Metrics Use Cases
Dave Dougherty: Yeah, and I think the coming back to a number of things that we've talked about, I think not only am I surprised that certain people still haven't gotten beyond, the answer engine portion of the AI tools. But also I'm still surprised that I get some of the questions I do at work from certain marketers.
What's the ROI of a QR code? Nothing. A QR code's a thing, and if you use it correctly, it can work well. But there's no ROI to a QR code. It's the thing that it's helping deliver. That's the delivery. But I bring up that because, again, we expect the robots to be totally perfect when the human inputs aren't even all that perfect.
Even when they're skilled, skilled marketers are skilled in a particular thing. I think the biggest thing for me, and Ruthi, you bring up the context piece of it, is and since our last episode conversation, I've been very cognizant of how I've been prompting. Because you're right. It's the here's what I want your help with.
This is what I'm expecting. This is what I've already thought of. This is what I want the output to be like, to help me process it. And that's a very different way of doing it than go do this thing, it because it's too open. And that's where I think a lot of people get stuck in their frustration with the outputs.
But, yeah. Anyway. Sorry. Lost my train of thought. Somebody jump in
Alex Pokorny: So Dave earthy, what's some things you want to get off your plate? For me, it was, I had an interesting conversation with someone talking about training for Google Analytics and Google Analytics is now has that an MCP. So there's a connection point to different Now
Dave Dougherty: right?
Alex Pokorny: a Google search console has the same and talking about training, I thought it was, do I want to spend time with Claude developing training or. Do I want to get it connected to one of these so that people can just ask it questions. And really my line was, I want to do the latter. I want them to be able to ask the questions because there's a main piece of metrics discussions, which is usually someone has a goal, but it's the wrong goal or they have a particular goal.
The way that they're going about it usually is the wrong way. The metric that they're trying to pull to prove the goal. the right metric. Or they need to really talk through or think through how would we actually find data that would prove or disprove this thought, right?
LLM Coaching Conversations
Alex Pokorny: Like of being like the end goal, which usually people come up with rather quickly, but that's not the way it works.
You have to go through step by step on building up to that point, right?
So that conversational piece, if that conversational piece can exist within an L lm, which I think it can. Probably with a little bit of help of, pre-programming on it, basically some preset prompts on it. That would probably be best training and it would save me a lot of time from a lot of random questions that I get, which I enjoy talking to people and I enjoy connecting with them, sharing expertise and the rest.
But I think it'd be better suited if people feel more comfortable that they could just ping this thing all day long at night, whenever.
Dave Dougherty: Right.
Adoption and Change Management
Ruthi Corcoran: I want to follow up with you in a couple weeks to see, or maybe a couple months to see are people using the thing. And the reason I say that is because I've created a couple of those now, and it's amazing how much I still just get the exact same questions, even though that's available. And then the piece that of course, I have, I want to figure out how to solve, is how do you, what does the change management look like for getting that behavior to start occurring? Or to shift or what's the. What's the slight tweak in the technical setup that needs to happen? Where you first look? Is there an agent? Not, is there a person or has somebody already shared this with me?
That's what I've already run into, where there's been a couple times where I've just created a co-pilot agent.
It's based on some series of documents and it's Hey, this is available to you if you have questions or you want to explore this further, here it goes. And that's like being like, here's my 20 page report. And in the way that people have responded, like sometimes you get an, oh, cool, how do you do that?
And, or that's interesting. And that's oftentimes as far as it goes, I'm curious your thoughts on just interaction alone.
Meet Users Where They Work
Alex Pokorny: I think it'll have to be wherever they are. That's going to be the main thing, is getting the basically company tool that they're already using to now have a new tool, new editability. And then for people to slowly recognize that it has that ability of trying to actually change them to learn something new come to a new place.
Or remember that there is a tool located at here. Remember from the one email that was sent once. Probably not. Instead, they
Ruthi Corcoran: It's like links to SharePoint sites.
Alex Pokorny: right?
Ruthi Corcoran: bookmark those.
Alex Pokorny: And people always just search again for it. Seeing that too, of a search gets better, then people bookmark less, right? So going to find it the way they're going to want to.
Dave Dougherty: Yeah. Part of that's a context problem, right? The larger the organization, the more the SharePoint, the more the the more you can get away with. Yeah. But just tell me. Whereas with the smaller orgs, build fast and break things because you can 'cause you're sitting next to the person who's technically in charge of your department.
They're right there. You don't have to do two weeks of trying to get on the calendar and being bumped and moved around and, so I think even structurally and org design wise, this is why for new technologies, new processes, smaller orgs have a way better ability to handle these things.
And. Okay.
Claude as Project Manager
Dave Dougherty: So the thing that I lost my train of thought on was like I feel like I'm finally getting to the point with cowork where the technology can finally keep up with my ideas because for 20 years I've been sitting and thinking I just need like a personal assistant. Because with all of the stuff I was doing at work and all the side stuff that I had signed on for between, coaching and the podcasting and, this other project that, that I'm talking about, and then my creative writing, it's just I am lost in the to-do list right now.
And I so I just, I created this co, CEO co, whatever, and just said. Here are all the links to everything. Here's the what I've been trying to do from 2025 into 2026. Here's what I've already outlined for my goals. I need you to be a project manager and I need you to help prioritize what I should be doing and here's all of the elements.
And I took an evening and just. Talked out all of the things that I'm trying to do, and it created not only tickets that could automatically be imported into the free Asana account that I have to do that, but it also created that little, tool within claw for these little cards so that when I log into that particular chat thread, I click on a card and it will automatically prompt itself.
Around that particular thing that I said I need to get done so that it has the context for me to be able to get it done. And then I just type in, alright, I did this. So like Alex, when you talked about setting up analytics, that was one of my things. Okay, I've been in this website migration for a while and I finally got to the point where, oh, okay, yeah, I should set up, GA four.
Search console and all of that. And it talked me through a workaround for index now on Squarespace, which was weird. I didn't know it could do that, but it said, here's the step by step. This is how you do it. Here's how you upload a file as a link. I'm like, what? And I was able to get it.
I was able to get it done. It was unbelievable. And it again, it was, like within a quarter's worth of basketball. Half hour. It was not that much. So it's been really cool because then also I don't have to keep track of that. I just log in and say, Hey, what were we working on last time?
Where should I start now? I've got X amount of minutes and it'll recommend. Here's something that you said you wanted to do. It'll take about this long. Let's get it done. So now I have a project manager, I have a research assistant. I have, all of these other things. It's dang, this is amazing.
You.
Enterprise Tooling Reality Check
Ruthi Corcoran: I think I, I just learned like 30 minutes ago that Claude might be available in copilot. So unfortunately I didn't get a chance to look into that before we met because some of my. My thoughts and reactions at the moment are just like, God, this sounds awesome. Really cool, Dave. Love it. I have zero personal projects for which I would use this.
That's just not where my hobby space is and all of my time is at, or most of my time rather, is within the workspace. And I know there's probably people listening balking going, ah, why don't you have a side hustle? Why aren't you doing all these things? And the reality is, because I'm a mom, we a homeowner.
I have a full-time job, we podcast, I've got friends, I've got family. There's only so much time in the day. I listened to this and what you described, and I'm like that's awesome. That's fantastic. How do I port that over into my work world? Which is all enterprise?
all enterprise. I got co-pilot, so maybe co-pilot's going to, in the next couple months, take a few, a few sort of additional steps into this. of ability to tap into other applications. But at the time being, for the time being, I haven't seen that as much. and so where my thought goes is how, come on, Ruthi, you can innovate, you can think a little bit differently. Are there ways in which I can start taking the approach that you're taking, Dave?
Which is, setting up the little small things that build on each other so that you can start creating that complexity and the ability of something like. An agent to do this when it becomes available within something like copilot. and I think right, being able to have a research assistant, have a project manager and do it in a way where it's decreasing the amount of time I'm spending on a given task or the next given task, especially the tedious ones.
Yeah, of course. That's hugely valuable. I'm not just, I'm just not sure how I get there with the current set of enterprise tools I have at this time.
Dave Dougherty: That's a good point too, because like I am. If you were to try to talk to me about what modern or what the current TV shows are, I have no idea. 'cause I just don't watch 'em. This is what I do for fun. I explore the ideas that come up in my head. That's just what I've done. That's what I do. That's what I get a lot of joy out of.
Dave Dougherty: That being said, even for the things that I want to do really well at so I coach my son's basketball team, right? And I went and I started finding all of these genuinely good coaching resources like from USA basketball, like team USA basketball. They have this great coaches corner kind of thing.
So I created just a. Notebook LM of these practice things for, here are the levels that I'll be operating in with this team so that I could have basically an assistant coach to help me with rosters. 'cause I also had to figure out equal playing time for nine kids every single weekend. And it would take me at least half an hour per game sheet, sometimes longer to make sure that every kid gets a start, every kid gets equal playing time.
And that we're also. Matched against the other team, right? Because it's one thing to do equal playing time, but then get smoked because you don't have a good balance of players on the court, right? It was an interesting like practice helper too, to be like, okay, thirteen's doing well, missed on these things.
Or how do I help so and so with? Focusing on this particular aspect of the game. And it was such a good time saver and it was something that I want to get better at. So I want to put in the work in, and I'm still researching on my own, like it's not a complete crutch, but to Alex's point, do I want to sit down and spend three hours doing rosters and practice planning?
No, I don't have time, but I do want to have enough time to develop something where the kids will have a good experience. They'll grow and we have a shot. And it worked wonders. It worked wonders.
Ruthi Corcoran: Yeah. And I that rings with some of the things I described either in the last episode or the one before in terms of, okay, I do gardening.
Gemini was super helpful in figuring out, okay, I'm going to grow a bunch of new plants this year. How am I sequencing, how am I making sure? And in terms of just time cut down and creating the timing charts. Yep. That was good. I also notice in listening to, to what you're describing, part of the effectiveness of some, of, some of cowork tools and using LLMs are predicated on the types of activities you're doing. And. And some of them work super well for saying, yep, we can cut down the amount of time spent.
And some of them, it's just not relevant.
And that's a good sort of thing to keep in mind of I don't know, as I'm trying to figure out where and how I'm shifting my day to day, and this is more in the workspace, just being conscious of there are going to be some areas where. It is just not suited at the moment.
Maybe it will eventually, but right now it's just not where it's at. And there's going to be some where it's yeah, you need to start thinking different or just test her out and try new things.
Task Agent vs Creative Partner
Alex Pokorny: I think that's kind of the context too. You think about how you're using these tools and one way it sounds like you're hiring an outside agency and you're giving them this piece of information, and then they go off and go, then they come back later and say, here it's, that's like giving an intern a task,
There's a big difference between that and an enterprise wide AI tool that has access to everything. And then you could say, Hey, what's our biggest three biggest competitors right now? And you would have some understanding of it and what's some, key marketing differences that we have for all of our different product lines versus their products or something like that.
You could really go in depth on something like that. Whereas right now you're more of, you'd have to provide a lot of that information so that it has now a better understanding of what's going on. So it gives you a more complete answer.
to your early point, Ruthi too, of, Dave asking for this in-depth strategy is knowing to ask that is such a different point you have to be in such a different mental space that you have to be in. Then stumbling around with this idea, where should I go next?
And if I had gone that direction, I don't think I ever would've thought of, Dave, your process of through the financials of how they grew
I would've looked at how they are today,
Dave Dougherty: Right.
Alex Pokorny: have said how do they grow and how can I, again, comparative to this idea, how would it, or not, or what would be the differences and. I wouldn't thought of that. So that's the other piece too, of like you have, you don't really have a creative partner, a coworker. got a task agent
and if there's things in life that are down to that level of task agent kind of stuff. Yeah, that makes sense. Especially if it's, able to be handed off entirely.
Dave Dougherty: All right.
Lawn Care Agent and Side Hustle
Alex Pokorny: We'll, maybe a follow podcast, but, I'm hoping today to build one of my first CLA agents personal use. and it's going to be tracking weather changes for lawn care. And there's certain
Dave Dougherty: Okay. No, I'm sorry. I shouldn't laugh, but that is,
Ruthi Corcoran: it.
Dave Dougherty: yeah, that such an Alex thing. Congratulations.
Alex Pokorny: Scientifically taking long care versus, the average person. Yeah, I know.
Dave Dougherty: Speaking of something I would never do. Yeah.
Alex Pokorny: When I get these emails that are timing emails for, okay, this is when you want to put this down now. And it all depends upon how the weather has been in the area, changes things by weeks.
And I also have. Pretty basic idea of how I want to handle things from a local store. So I want to have also an update notification two weeks beforehand, because it'll probably take me about two weekends to finally do the errand to go buy the stuff, right? So it's I want to have this thing, build this thing out, keep checking this thing for me so I don't have to care about it. And yes, I already get the emails and the alerts, and quite honestly, what I'm doing right now will work just fine, but. It'd be nice to be able to hand that thing off. I don't want to do it and tell me when to do certain things for lawn care. Like honestly, I don't care that much. I want the output. I don't want the work or the input part. Like I don't find any joy in setting up a process for that. find any joy, to be honest, putting down fertilizer or something. I just want the output. That
Ruthi Corcoran: Alex, if we could con combine our ideas, I think we'd be onto something. I, perhaps somebody already has this in the market. Dave could do the research for us. But we could, combine our ideas. You've got the lawn care angle, I've got the like flower and we could give people here's your garden lawn in a box and all your,
Alex Pokorny: Home
Ruthi Corcoran: and connect it with and just have things shipped. Love this plan. Let's do it.
Alex Pokorny: More green in people's lives for the things that're not confident in growing, which
Ruthi Corcoran: Ah,
Alex Pokorny: helped me figure out which plans to buy for my home office. 'cause it's weather dark and maintenance is low. So
Ruthi Corcoran: very nice.
Alex Pokorny: do it. There you go. Now Ruthi has a side hustle.
Ruthi Corcoran: I've got a side hustle guys. Mission accomplished.
Dave Dougherty: You just gotta find your thing ultimately. Yeah.
Ruthi Corcoran: True and as long as I can do it all, voice to text so that I can be gardening while instructing Quad to do all this, that'd be great.
Dave Dougherty: that is, yeah, that is a whole other episode of how you input things. Yeah.
Model Versus Model Workflows
Dave Dougherty: Also for next for next episode I've started pitting one model against another. So because. Because I'm trying to do this like business research, play around with the idea kind of thing. I'm taking the output of one and then putting it into another and then ramping up the critique factor.
And so here are all of these models and here's all these hand, these handwritten notes from my business classes from back in the day. So use these formats and these frameworks and whatever else. To find holes in the ideas in this other one, and how would you change it? So getting these things against each other to then, fill in gaps or get pointed into another sort of research direction has been, the next thing that I'm starting on.
Wrap Up and Call to Action
Dave Dougherty: But anyway, that's more than enough for one episode. Thank you for sticking with us this long. Please go check out enterprising-minds.com where you'll get all of the transcripts and all of the newsletters once you sign up for that. We're also on substack for our newsletters and podcasts.
So go check that out. Subscribe there. And without that go play, even if it's not for your company. Do it for your lawn if that's what you're into, or meal planning or whatever else. Have fun. We'll see you in two weeks and take care. Bye.
