How We ACTUALLY Use AI Daily: From Figma to Fighting AI Slop | Utilizing AI Episode 30

With so much happening in AI, it's helpful to stop and consider how IT professionals are actually using it in their day-to-day work. This episode of "Utilizing AI" features Tech Field Day delegates, Jim Jones and Jeff Ellough, discussing the ways they are and are not using AI every day. Welcome to "Utilizing AI," the podcast focused on practical applications of artificial intelligence from the Futurum Group.

Every Wednesday, we explore news and use cases of the ways in which AI is transforming enterprise IT and the industries it serves. I'm your host, Stephen Foskett, President of the Tech Field Day business unit here at the Futurum Group. Before we dive into this discussion, let's meet who's on the panel today.

Hi, everybody. I'm Jim Jones. I'm a product architect at 11:11 Systems, cloud service provider based out of New Jersey.

info or pretty much everywhere on social at KoolAidIT. And I'm Jeff Ellough. I'm an AI lead at Purdue University.

com or LinkedIn or any place I would be. And of course, we'll find all three of us at Tech Field Day events. Today, we're doing something a little special here.

We've invited on a couple of our Tech Field Day delegates who are deeply involved in practical applications of technology generally to talk a little bit about how they're using AI, and how the technology has affected them in their real daily lives. So, Jim, let me start with you. Talk to us a little bit about your job, as well as how you're using AI.

Yeah. So, here at 11:11, as I mentioned, I'm a product architect, and in our speak, that means I'm more the technical side of our product process, PDLC, if you will. So, that starts for me in the early R&D stages of, here's this thing that we're going to do, which is very Frankenstein-like of mashing together the APIs of four or five vendors at once, through the productization process to, on the backside, doing a lot of the marketing, public speaking, content creation, et cetera, that goes along with it.

For me, AI has become very much so a daily driver in about, I don't know, four different ways, really. A lot of our ideation work, especially when it comes to UI work, starts in Figma Make. If you guys are familiar, I'm sure you've heard of Figma, but Make is their AI component to it, to where you can natural language talk to it, give it a starter and say, "Here's what our UI mock-ups look like.

Go create this widget, create all this," and it helps us very quickly proof of concept for dev and create the screenshots and all that kind of stuff they need. Going into the documentation side of it, there's a lot of different platforms. We historically now have been using Glean.

There's a little bit of Claude that goes in there as well. If you've ever worked in product, you know that it's 99% documentation and 1% we're actually going to make the thing. So, using AI platforms to go through, do that PRD creation, all the core team requirements.

Who should actually be touching this thing? Doing the research on that. Market intelligence, which is also a massive component of what we get into in product.

And so, anyone that's making a thing that's not just Jim at night with Claude code just going crazy, nine times out of 10, you've got to make the case for, is there going to be an ROI for my company to do this thing? So, using it as a capability to go do what is normally weeks of work of what are all our competitors doing? How can we do it, how can we differentiate, and when are we going to get some money back from it?

That's becoming a couple hours conversation at this point. Aside from that, it's just a lot of proof of concepting through code as well as anything is really where that is hitting the road for me professionally these days. On the higher ed side, this is hitting us like a sledgehammer because the students are coming in, they already know all the tools.

We have faculty who are brand new to actual researchers in AI. So we run the gamut of every type of user and every skill level, and we're literally just trying to get our arms around it, trying to come up with ways to avoid academics devolving into an AI slop-generated homework fest. It is something new every single day.

And I wish I could be more specific about it, but you go off into different tangents very quickly. The number of tools that are available now, and especially with the agentic stuff that has come out, has exploded so quickly and across every possible facet of academic life that it is almost impossible to even set up guardrails before they've been breached. You just can't.

They're already using it in every possible way. You sit in the commons area, and you hear them talk about Claude Cowork and how they had Gemini do this whole process for them in their setting up for a presentation or whatnot, group work that used to be done by one person. Now you've got four people who are well-versed in using AI to come up with some pretty unique and amazing presentations.

Well, let's just say that it's gone a lot further than we expected and a lot faster than we expected. We are trying to build tools to automate daily life, things that are difficult for people to do. Because we're a university, there's a lot of just general administrative work that nobody actually wanted to ever do.

And we're providing the tools so that people don't have to do that. They can spend more time with students, spend more time just doing the things that make student life better. And that's what our group is trying to do, as quickly as possible, I might add.

On the side, I am also involved in both media and filmmaking work, and so bringing AI tools into that workflow and all of the problems that are associated with that, those consume a bit of my time as well. Yeah. And I think we've done some talking in the past about Futurum, but I think, just to sort of provide a me too here, Futurum is a very aggressively AI-forward company, right from the very top.

Daniel Newman directed us all to leverage this technology, and I've always been very pro-technology, generally, pro-automation. And so I've found that our team, the Tech Field Day team, uses AI tools almost everywhere. But that being said, one of the things that struck me is that as we've become more familiar with the technology, and as we've really started leaning into it, we've found that we, as strange and paradoxical as it sounds, it's more important to have it be a tool used by people than it is to have it be an autonomous system.

So even me, who is an aggressive process automation freak, I would say, I find that I use AI as one of the steps in the process more than as the process. " or something like that. But all of my attempts to use it as an autonomous system, basically as an agent, haven't really worked out to this point.

So, let me throw it to you all. Jim, I guess we'll start back with you here. To what extent are you using AI as a tool for yourself as opposed to as an autonomous system?

Yeah. So, one of the other functions that I do both here at work as well as with a local community college that I work with a lot is kind of giving guidance as to other people that aren't maybe necessarily as technical of how, when, and what to use AI with. And the biggest thing that I always lead with is there is a lot of power in being the SME.

Frame your concept, frame your use of, I am always going to know more about this, whatever this niche subject is that I'm working with the AI with, than the AI model itself will. And so as you instruct it to do the thing, to kind of lead with that in mind. So yeah.

So I am definitely using tools. So when I work with the university, recently did a presentation there on cybersecurity. What I did, I let Claude create my entire PowerPoint presentation, but it was baked down to within every single line in an outline of here's how this content works with exactly what I wanted, and then very explicit prompt of, okay, here's what I want you to create, here's how I want you to use this outline, and here's how I want you to generate the output.

Effectively, there's a lot of content that you see going around today, and Jeff, I think that's probably a lot of what you're running into a lot of. It's a whole lot of one-shot. " Whereas the more you put into what you actually tell the AI to do, even if it is a one-shot, that one shot can just as well be a thousand words as it is 100 words.

The more you tell it... the better that output is going to be. And that's like my North Star in terms of my interaction with the AI and how to work with it.

Jeff? One thing that's always true is that I have a billion emails at any one time. So I set up an agent to check that every day, find out what am I missing, what did I forget to do last week while I was at AI Field Day, things like that.

These are generally automated systems. I'm very aggressively automation. That's always been my thing.

I do try to get the interface out of my way. However, where AI has sort of changed the way I view my own job, I've always been a very good programmer, but I hate coding, period. So I stay away from it, let other people who enjoy doing that, do that.

But, it is now so easy and I don't feel disheveled or whatever after a few hours of coding with Claude or whatnot. And so it has actually brought me back into the fold on development, and that's actually been a nice change, because I wish that I could have spent more time doing development. And now I feel like I can actually contribute meaningfully to some development work.

Another thing, so we generate tons of content, video content, textbooks, things like that. And, I have always wanted to have a library of the content that we've created. We've created tens of thousands of videos of faculty delivering academic content.

I've always wanted to, but, in fact, 15 years ago, when we started this little project of doing all these videos, we said at some point, we'll have a library. We'll be able to search and have a student who's in one particular discipline get information about the thing that they're learning about from someone in a completely different field, right? The same topic just explained completely differently because people learn differently and different ways of explaining it can work for one student or another.

So AI is helping me build that library, that cross-linking. So, we're using it in strange and unusual ways. It's not like it's something I pull up every five minutes to write an email.

That's just not how I use it. But for doing these things that would otherwise be impossible for a person or a team to do. " Yeah, we are seeing that, especially on the remedial stuff.

So, you went to college, but your math wasn't great, right? And you need some amount of... So instead of going into a remedial math course or whatnot, we are seeing people proactively use AI to teach them concepts that they weren't getting.

And even in elevated classes, there's all sorts of individual sections you might not understand, and they will almost immediately turn to ChatGPT or something like that to explain the topic differently. Because that's the one thing, is you can have a topic explained 100 different ways, and for you specifically, one of those ways is going to click, whereas everything else doesn't work. And so the more you ask, or if you had 100 AI tools and you ask one of them, one of them is probably going to explain it the way that you will understand it, which is, that's a nice tool to have.

Yeah, I ask because I'm actually going through some of this myself right now. Being in tech certifications, I should be old enough that certifications don't play a part anymore, but I just can't quit the habit. So I'm currently, we're a big VMware Broadcom partner, so I'm working through the new VCF certification stuff, and the certification guide for the VCP, just the base level was like 900 pages.

" And what I'm finding is that I can tailor the education to where I'm at with my career, with life, et cetera, and get to an outcome that actually works. I find that to be very helpful. Yeah, one of the ways that I've found a lot of people using LLMs most effectively is to do what you're describing as a collaborator, essentially somebody that you can sort of work through things with.

I'm saying somebody, of course, it's not a real person, but it's a very good tool. If LLMs didn't exist in our consciousness the way they did, and I had a tool that you could literallyput things into and have it kind of churn through them with us, that would be valuable right there. " And I've been doing that quite a lot.

" And that's not a way for it to generate information. It's a way for me to generate information and it to assist me in following my thoughts. And in a way, that's as well what you were just talking about, Jim and Jeff, the idea of using AI as sort of a personal tutor.

It's not a tutor that knows everything. It's a tutor that can help you explore a topic, right? Yeah.

So that's pretty much the approach that I've been taking with AI. I don't want to really delegate anything to it. I think we all have our own trust journey with artificial intelligence, and that's kind of the place where I'm at right now, that it's very much so a distrust but also very much so verify kind of capability.

But I want to enable it to, like you said, Steven, be able to help me focus my thoughts. If you've not used it, and I don't know if it's imported to Codex or other things yet, the superpowers skill for Claude is amazing for that collaborative approach to go at doing things, whether it's programmatic or whether it's creating documentation. What I find is if I'm wanting to do new things, I'll go through and create a feature document of, here is every thought, this is effectively the brain dump that I have on this general concept.

Write it in whatever form that I want. I'm naturally an outliner. " And it natively, all by itself, knows, okay, parse this thing, get the general information, go through, ask concerning questions, and then let's come up with a plan collaboratively together that actually outputs as a markdown file that can then be edited before you ever implement the first thing.

It's really powerful, really amazing with that. I don't know if you guys have had a similar situation. I like putting AI in the role of the person I'm talking to.

So pretend you are CTO of something something, and then run what I intend to say through the AI to see how it would respond and whether or not it would be perceived the way that I expect it to be perceived. Not that it'll be right, but it will often say something that will trigger me to change my maybe specific verbiage or use a less loaded word, things of that nature. It's super helpful for reframing my thoughts into a way that would be more beneficial for others.

So let's talk a little bit, I guess turn the page and look more toward the future here. What I would love to do is think about, you all have a pretty good understanding of this. Jeff was just at AI Field Day with me.

Jim, I know that you're keeping on top of the technology as well. Realistically, how do you see yourself benefiting from this technology in 2026 and 2027? How do you see this coming?

Not what do the companies wish that you would do, but how do you think you'll do? Jim, let's start with you. Oh, goodness.

Put me right on the spot. So what I'm finding is that it is doing a good job of helping me take very small form, call it tweet-sized ideas, and fleshing them out into something that's a little more meaningful, a little more impactful, a little more corporate ready. And again, to that collaboration, and Jeff, to your conversational kind of approach, that it makes it very easy to, before it ever escapes your own look, be able to shop that around to different people or different personas and be able to make use of that capability.

That's where a lot of my focus is today. It's not so much I'm going to go have it build a whole Git repository full of code that's going to be my first SaaS application. It's more, how can I take the things that are already kind of percolating in my head and turn them into something that's more consumable by others?

Yeah. I'm seeing the same thing. The issues with students and faculty and how they're using it, those are my more concerning directions because we're really trying in the higher ed space to make sure that students are learning as opposed to simply having the AI do their homework for them or think for them, and we're seeing a lot of changes.

So at least in higher ed, we are going to see over the course of the next year big changes to how we teach, how we test, how we go about assessing people and their skills, and especially teaching people how to use AI in a way that is constructive and builds on their own skills rather than replaces them. Yeah. I think that to echo what you both are saying, this is a tool, and it's a tool that makes us do what we do better.

I think that where it goes wrong is when people look at it-as a replacement for human thought, human expertise, human effort. It's just not. It is incredibly good at some things, and I think that many of us are finding that it really does make us work more effectively.

I love what you said there, Jim, about communicating more effectively with different people. Jeff, you mentioned that as well. Set up as a persona, let me figure out how I can talk to this person better.

That's a really powerful use of a chatbot as opposed to let me let this thing loose and see what it'll do. And I don't know that people are really going to head in that direction, especially now that the true cost of a lot of this technology is coming in. I don't want to derail the conversation or drive it into another direction, but another thing that we're watching now is a lot of these cloud LLM companies are starting to charge for token usage, and that's really surprising people about just how much resources they're using, they're consuming to do things that may be dubious.

So thank you so much for sharing your perspectives here. I do encourage our listeners to continue listening to Jim and Jeff and the rest of the Tech Field Day delegates. You can find them on the Tech Field Day website.

You'll find them in our discussions and our videos, the Tech Field Day podcast, and so on. But before we go, I want to give each of you a chance to tell us a little bit more about where we can connect with you and continue the conversation and find your thoughts on AI and other technology. So I guess we'll just go in the same order here.

Jim, where can we continue the conversation with you? info, and then all the socials are linked from there as well. com or LinkedIn.

Excellent. And as I said, Jeff was just at AI Field Day. We did a great roundtable discussion about this.

Jeff's appeared on our podcast lately and so on. So do keep an eye on Jeff and Jim and the rest of the Field Day delegates. " If you enjoyed the discussion, please do subscribe on YouTube or in your favorite podcast application and consider giving us a rating and a review.

This podcast was brought to you by the analysts and the experts at the Futurum Group, where insights meet AI. ai, the "Utilizing AI" YouTube channel, or the Techstrong TV app. Thanks for listening, and we'll see you next week.

Now hang