Mastering the Art of AI Prompts

In this episode of Prompt and Circumstance, Tom Adams, Mike Richardson, Mark Redgrave, and Ryan Niemann dig into the art and science of prompting. They share stories of building real tools without coding, debate the importance of prompt discipline, and outline frameworks that every CEO should know as AI becomes core to business strategy.

The crew covers:
- Tom’s no-code apps built in Warp for personal healthcare tracking and in Claude Code for field reps.
- Mark’s perspective on compute costs and why clarity in prompting equals clarity in leadership.
- Ryan’s frameworks—from “Act as a / I will / You will” to the six-component model of prompts.
- Mike’s experiments with Plaud AI recorders and his new Surface AI laptop.

They close with doable actions for CEOs: treat prompting as a core business skill, solve small recurring problems with AI, and form internal councils to experiment with AI capabilities.

----------------------------------------
Show Links
----------------------------------------

TOOLS & PLATFORMS
- PromptWorks Pro (six-component prompt builder) – https://promptworks.pro
- ChatGPT (OpenAI) – https://chatgpt.com/
- Claude Code (Anthropic) – https://docs.anthropic.com/en/docs/claude-code/overview
- Microsoft Copilot – https://copilot.microsoft.com/
- Perplexity AI – https://www.perplexity.ai/
- Warp (AI terminal) – https://warp.dev
- Plaud AI (recorder) – https://www.plaud.ai/
- Whisper (speech-to-text by OpenAI) – https://openai.com/research/whisper


CONCEPTS & FRAMEWORKS
- Prompt Engineering – https://www.promptingguide.ai/
- “Act as a / I will / You will” structure
- Iterative prompt refinement (“Improve my prompt” technique)
- Agentic AI and conversational vs. product prompting – https://en.wikipedia.org/wiki/Software_agent
- Six-Component Prompt Framework: Role, Task, Context, Reasoning, Format, Success
- “Think harder about this” (leveraging GPT-5 reasoning toggle)
- "The Frontier Firm" How Intelligence on Tap will rewire business (from Microsoft’s Worklab research and Work Trend Index Annual Report)

----------------------------------------
Key Quotes
----------------------------------------

Mike Richardson
- “If you are able to keep up with me, you’ll be kind of keeping up with them … I’ll do my best to hang on to their coattails.”
- “I got myself one of these Plaud AI recording devices … I’ve just started experimenting with keeping it on all day.”
- “I’ve had more glitches to work through with a brand-new laptop than ever before.”
- “Doing conversational prompts in ChatGPT, Copilot, Claude … that’s good training. We’ll reap the benefits of that when we move to agentic stuff.”
- “Whether you’re using Plaud AI or Copilot … I urge you to get in the habit of using AI companions for every meeting you have.”

Mark Redgrave
- “Before 2010, AI compute doubled every 21 months. Now it’s every 5.3 months … one estimate has the cost surpassing the entire US GDP by 2037.”
- “If you can’t describe what you want, maybe you don’t have the clarity you need … prompting becomes a proxy for whether you understand what you want.”
- “You should be prepared to sit and take time to write a great prompt … I force myself to slow down, even when I want to just hit enter.”
- “Prompting is gonna be table stakes … every team member is going to need to know how to prompt effectively.”

Ryan Niemann
- “Act as a … I will … You will.”
- “If you provide a prompt and then say, would you improve my prompt before proceeding? … it will.”
- “You can also say, ask me three questions before you proceed … it really improves the results you get back.”
- “Role, task, context, reasoning, format, and success — those six components can help improve your prompts.”
- “Try different LLMs … put the same prompt in ChatGPT, then Claude, then Perplexity, and see the different styles you get back.”
- “Think harder about this … you can trigger that in GPT-5 to get deeper reasoning.”

Tom Adams
- “Basically what I built without knowing how to code … was an app that allowed her to voice her challenges … it transcribes it, sends it to AI … and then we’ve been actually taking that to our medical appointments.”
- “We got this problem, like my field reps don’t say something when they see something. And I said, let’s build an app called When You See Something, Say Something.”
- “I don’t know how to code … and yet I’m able to come up with this stuff in hours, maybe even minutes.”
- “Often how we have to converse with a person that we’re asking to do a specific thing … that’s what’s so profound about this — teaching a machine in the way we’d teach a human.”
- “I literally spend all my day just talking … prompts all the time.”

----------------------------------------
Doable Actions
----------------------------------------
- For CEOs (Mark): Treat prompting as a core business skill. Start building capability across your teams.
- For Practitioners (Tom): Don’t chase big AI projects—solve a small recurring pain point with prompts or agents.
- For Leaders (Ryan): Create an AI council or small working group to explore use cases and grow internal expertise.
- For Individuals (Mike): Start using AI meeting companions daily to capture action items, tone, and productivity insights.

----------------------------------------
Meet the Crew
----------------------------------------
Mike Richardson - Agility, Peer Power & Collective Intelligence
https://mikerichardson.live/
https://www.linkedin.com/in/agilityexpertmikerichardson/

Ryan Neimann - Software CEO & Board Operator
https://bob3.pro/
https://www.linkedin.com/in/ryanniemann/

Mark Redgrave - Agility, People and Performance
https://www.shift-transform.com/
https://www.linkedin.com/in/mredgrave/

Tom Adams - Executive Coach, Advisor & Trail Blazer
https://tomadams.com/
https://www.linkedin.com/in/tomadamscoach/

Creators and Guests

Host

Mark Redgrave

Agility, People and Performance

Host

Mike Richardson

Agility, Peer Power & Collective Intelligence

Host

Ryan Niemann

Software CEO & Board Operator

Host

Tom Adams

Executive Coach, Strategic Advisor & Thought Partner

What is Prompt and Circumstance?

It’s human-friendly banter about code, culture, and CEO reality checks—served up by Mike Richardson, Ryan Niemann, Mark Redgrave, and Tom Adams. No jargon. No hype. Just real talk from four guys who’ve seen it all, and aren’t afraid to say what everyone’s thinking.

[00:00:28] Hey. everybody. Welcome back to another episode of Prompt and Circumstance, uh, the podcast that you've gotta listen to if you are trying to illuminate the pathway forwards with your AI strategies, so you are wandering around in the darkness a little bit less than everybody else. We are here to have a reality check for executives and CEOs.

[00:00:51] mike-richardson_1_08-22-2025_100238-1: If you wanna know more about us, go to prompt and circumstance podcast.com to read our bios. But let me tell you who we've got here today. We have the amazingly practical Tom Adams. He blows our, he blows our minds when we think it. He produces an app for it within about 30 minutes using.

[00:01:14] tom-adams_1_08-22-2025_130237: to code,

[00:01:15] mike-richardson_1_08-22-2025_100238-1: Without you, without knowing how to code, using ag, vibe, this, that, and the other.

[00:01:19] We'll find out more about that later, everybody. We have the eternally optimistic Mark Redgrave, or at least on a good day. On a good day, he's eternally optimistic. Not so much in the first episode. If you wanna know what that's about, go and listen everybody.

[00:01:35] mark-redgrave_1_08-22-2025_100237: I had a, good chat to

[00:01:36] myself after the last one, so I'm gonna try and try and be better this

[00:01:39] mike-richardson_1_08-22-2025_100238-1: We have the ever underestimated Ryan Neiman, everybody. If you read his resume and you understand his experience and you understand his perspectives and where he's headed and where he's going, not least of all, as you're about to find out in this episode, it's just amazing.

[00:01:58] tom-adams_1_08-22-2025_130237: Mind boggling. Not even worthy. Not even worthy.

[00:02:01] mike-richardson_1_08-22-2025_100238-1: And then yes, you have me, Mike Richardson, uh, the slow guy in the room bringing up the rear. If you are able to keep up with me, you'll be kind of keeping up with them. Well, kinder, we'll sort of do our best to hang on to their coattail. so LIS, listen everybody, generally speaking, from episode to episode.

[00:02:21] We're gonna sort of open with, you know, what's the new news that everybody's been focused on in the last couple of weeks? Uh, what's the new stuff they're reading, hearing, seeing, playing, with, working on just the new news, everybody. We'll finish with what are the doable action steps that we recommend that you take next listeners, just to find your pathway forwards. and then we'll always have some meat in that sandwich. and the meat today is gonna be so what is prompt engineering?

[00:02:50] How do we think about it? How do we look at it and how do we blaze the trail with getting better and better and better at this thing called prompt engineering? But we're gonna start with what's new everybody.

[00:03:02] What have you been playing with, working on seeing, hearing what's out there? And we're gonna start with Tom Adams.

[00:03:09] tom-adams_1_08-22-2025_130237: All right. Uh, I promised last time that I was going to be working in terminal, and I don't even know what terminal is, but I promised I would be u working in terminal and terminal's, like the thing that, that has like a prompt, like it's just a box there. To get into the back end of your machine, and I've been playing with this tool called Warp and uh, my wife's currently having, uh, some health challenges.

[00:03:33] And what I decided to do was ask Warp. In terminal to help me build an app. And basically what I built without knowing how to code and without really knowing what this thing was doing on my computer, was build a app that allowed her to voice her challenges that she was having, what her experience was in the current moment.

[00:03:53] 'cause she was struggling to write it all the time. And so she pro, she would voice it, this thing transcribes it, sends it to AI to. Really, uh, restate it as sort of more, um, specific issues that are happening and then documented in a spreadsheet that has date, time, um, specific issues. And then we've been actually taking that to our medical appointments to say, this is what's happening, this is what's actually

[00:04:18] going on.

[00:04:19] So that's been really cool. Um,

[00:04:21] mike-richardson_1_08-22-2025_100238-1: What did we say? Everybody? Amazingly practical. That is Tom Adams through and. through.

[00:04:25] tom-adams_1_08-22-2025_130237: Yeah. Like just doing that. And uh, and this week I decided, uh, I'm gonna try and keep up with Ryan and build a paid app to figure out if I can build a paid app without knowing what I'm doing. And so, I was talking to a coaching client and they said.

[00:04:40] We got this problem, like my field reps don't say something when they see something. And I said, let's build an app called when you see something, say something. and so what I'm doing is a similar kind of concept that I did with my wife's, which is build an app that if they see it, uh, opportunity with a client, they press the opportunity button and then they talk about it and then they

[00:04:59] take pictures of it and it submits it and goes to head office.

[00:05:03] So,

[00:05:03] mike-richardson_1_08-22-2025_100238-1: Just remember what Tom said earlier, everybody. He does not know how to code

[00:05:07] tom-adams_1_08-22-2025_130237: I don't know how to code.

[00:05:09] mike-richardson_1_08-22-2025_100238-1: and yet he's able to come up with this stuff in hours, maybe even minutes.

[00:05:14] tom-adams_1_08-22-2025_130237: So that, that's what I've been doing this week. And, when we talked about prompt engineering, I

[00:05:18] am going to talk about all the challenges I've had with that one because,

[00:05:22] mike-richardson_1_08-22-2025_100238-1: Okay.

[00:05:22] tom-adams_1_08-22-2025_130237: Claude code completely broke

[00:05:24] everything when I prompted it and it broke my entire system. And so that's been

[00:05:28] mike-richardson_1_08-22-2025_100238-1: Alright, everybody hold. Alright, everybody, hold your Breath.

[00:05:31] Let's find out what kind of mood mark is in today. Mark. Mark gra what? What's new on your world?

[00:05:36] mark-redgrave_1_08-22-2025_100237: I've had a great, I've had a great week, actually, Tom, if you, if you've made that much progress with, with your wife's health stuff, like, I'll just cancel my Hogue stuff and send the money to you. I'll just, because, and, and it's like, it's like a $95 million a month I spend at Hogue, so

[00:05:52] I, I could give that to you if that's, if that helps.

[00:05:54] tom-adams_1_08-22-2025_130237: Yep. Any, any little bit

[00:05:56] mike-richardson_1_08-22-2025_100238-1: so what? What have you been playing with? What have you been seeing? What have you been working on? Mark?

[00:05:59] mark-redgrave_1_08-22-2025_100237: I, um, I've had quite a creative week, so, um, you know, in response to a number of conversations from CEOs and leaders, uh, I've been, I've just been forced to get clarity on, you know, what some of the principles where a number of us come from around. Agility and adaptability is like, I think organizations are facing with this. The ai, onslaught is just forcing them to, rethink some of these principles that are foundational. So I, I've had a great week actually, but talking for, not, not in prompt engineering and in app development, but more in terms of like, the narrative that I think CEOs, uh, need to hear to clarify that.

[00:06:32] So that's good. I've got something I've gotta share though. This is so good. I dunno if this is right or not because to be honest, on LinkedIn, I read things that are quite contradictory. But this was, this made me like, this made me laugh. So it says here we're talking about the cost of compute, right? And it says, before 2010, AI compute doubled every 21 months.

[00:06:54] Okay? Now it's every 5.3 months. A GI will require computing power exceeding 10 to the power of 16 terra flops. Okay? One estimate the cost of that surpassing the entire US GDP by

[00:07:14] 2037.

[00:07:17] mike-richardson_1_08-22-2025_100238-1: and and again, mark, if we can translate into English, a terror flop is what?

[00:07:22] mark-redgrave_1_08-22-2025_100237: a lot. A lot. It's terra flop is Latin for a lot. A lot, but

[00:07:29] tom-adams_1_08-22-2025_130237: It's when it's so much you fall down.

[00:07:32] ryan-niemann_1_08-22-2025_120238: mark, didn't you, didn't you introduce the thought

[00:07:34] that every prompt requires a glass of water

[00:07:38] mark-redgrave_1_08-22-2025_100237: Yeah.

[00:07:38] mike-richardson_1_08-22-2025_100238-1: Oh, that's right.

[00:07:39] mark-redgrave_1_08-22-2025_100237: and, and I read this morning,

[00:07:48] Ryan, I read this morning Ryan, that that like a query right is takes the same power consumption as six seconds of TV watching. And that was Pres that was presented right? As a positive. It was like, it's like, gosh, aren't we, we're getting better. I'm like, holy crap.

[00:07:57] We, we saying every time somebody puts thank you into chat, GBT 'cause they've got something useful.

[00:08:03] watching six. Anyway, some of these numbers are mind

[00:08:06] boggling. So

[00:08:06] mike-richardson_1_08-22-2025_100238-1: Alright. I like,

[00:08:08] mark-redgrave_1_08-22-2025_100237: this week, so

[00:08:09] mike-richardson_1_08-22-2025_100238-1: I like Mark's mood, everybody. We're in luck this week. Excellent. Ryan Neiman, what have you been, looking at? What have you been working on?

[00:08:15] ryan-niemann_1_08-22-2025_120238: You know, I was introduced to an article called,

[00:08:18] the Frontier Firm

[00:08:20] mike-richardson_1_08-22-2025_100238-1: Hmm.

[00:08:21] ryan-niemann_1_08-22-2025_120238: and I thought it was absolutely insightful from providing a framework, a new glossary of terms. But it really frames I think, what we're feeling, an urgent sense to share. With, some others. Uh, it goes into how agents will be introduced into companies. the concept of being a boss of agents and how that begins to continue to develop, how there'll be an adoption of more digital labor, how there will be a need for a human. ratio. Like I often, I, I have felt this myself. I know Tom and I were talking about, you know, when you've got a number of agents working simultaneously, there's, there's a bit of cognitive fatigue and so there's some thought about what, what will be that ratio?

[00:09:13] And you find yourself as a, a manager, you know, managing a number of bots and being that human in the loop. Um, and the last thought in that was that there's this sense just in summary of. Intelligence on tap. And what that means from being able to, rapidly become somewhat of an expert in many cases in a particular field that you weren't very familiar with previously.

[00:09:38] And I, I dove in on that. I think it's Microsoft funded and researched. Uh, they mentioned that they leverage. Trillions of copilot, uh, uh, transactions. They have a very global footprint in in this study, uh, and also leveraged LinkedIn data in what they're seeing in both, uh, large organizations. Growing at, uh, I think they said year over year, 10%, but smaller firms in technology growing 20%.

[00:10:08] So a lot of exodus of, tech

[00:10:10] talent from big firms to, uh, small firms. It was a very insightful article.

[00:10:13] mike-richardson_1_08-22-2025_100238-1: Wow. Wow. Awesome.

[00:10:15] mark-redgrave-audio: Frontier Firm.

[00:10:16] Frontier firms are cool. I like that.

[00:10:19] mike-richardson_1_08-22-2025_100238-1: really cool. me the slow guy in the room. I was playing with my new toy, which is I got myself one of these applauded AI recording devices. that, uh, can record for up to about 20 hours continuously, and it's got like 20 days of, you know, standby battery life.

[00:10:36] And so I've Just started experimenting with that and been recording meetings and, you know, just keeping, trying to keep it on all day sometimes and seeing what I get out of that. And then through the web app, interrogating it at the end of the day, you know, Hey, how did I show up today? You know, any, any.

[00:10:55] Any sort of missteps anywhere where, you know, my, my tone or my presence wasn't how I want it to be. Those kinds of things. And then of course, easily getting, you know, um, summaries from different conversations that I've had, you know, Just in the flow of my day. So I'm really excited about that. I did get my new Surface Pro AI ready, Snapdragon co-pilot enabled.

[00:11:20] laptop to work, although bizarrely, I mean, you would imagine everything was plug and play these days, wouldn't you? Bizarrely I've had more glitches to work through with a brand new laptop than ever before, and I'm still, I still don't have all the wrinkles Mike, just because you got the Tim one, that's not the one. That's not the one I told you. anyway, anyway, they cost more than $12 for 50.

[00:11:47] I spent a lot of money, I spent a lot of money on this.

[00:11:49] tom-adams_1_08-22-2025_130237: Mike, get a

[00:11:50] Mac.

[00:11:51] mike-richardson_1_08-22-2025_100238-1: I, I, I, spend a lot of money on this laptop. Anyway, it's meat. In the sandwich time, everybody, we are gonna come to this question of what is prompt engineering? And what we want to get into here is how do we think about it?

[00:12:06] How do we look at it? How do we blaze the trail with it and how do we really start to refine and optimize and continuously improve the way that we are prompting? And perhaps, Tom, we'll start with you and then we'll come to Ryan, 'cause Ryan's got some really magical stuff that he's been working on. But Tom, can you, can you get us started down that path?

[00:12:26] tom-adams_1_08-22-2025_130237: Well, I, I'll start

[00:12:27] and I feel like we should go to Ryan and then I'll, I'll come back with some sort of

[00:12:31] mike-richardson_1_08-22-2025_100238-1: Okay. Let's do that then. Okay. Alright. Nice pivot. Nice pivot, Ryan.

[00:12:35] tom-adams_1_08-22-2025_130237: I, I don't, I don't mind

[00:12:36] that, but, but, um, but I, I just think

[00:12:40] mike-richardson_1_08-22-2025_100238-1: Mark's holding up a sign. Everybody saying, I'm here to learn.

[00:12:43] mark-redgrave-audio: I don't, I, I've, I've got so much to learn here, Ryan. This is like, this is money time. Like hit me.

[00:12:47] tom-adams_1_08-22-2025_130237: Yeah. Yep.

[00:12:48] ryan-niemann_1_08-22-2025_120238: All right. Well, I would put it in some context.

[00:12:51] tom-adams_1_08-22-2025_130237: Yep.

[00:12:51] ryan-niemann_1_08-22-2025_120238: say getting out of 2022 into 2023, I was looking for ways to prompt better, uh, I was finding that when I was talking to both colleagues as well as some family members, that there's just some education on, on how to, uh, improve, uh, the way that you are working with any particular large language model. Um, one thing to note. talking about a couple terms here that are actually contentious. You wouldn't think it would be, but prompt engineering does have, uh, some contingency that believes that's a really bad name. Uh, and, and contend it's not engineering at all. Uh, also vibe coding. Uh, a lot of coders take exception to vibe cutting, so I'm just gonna get that on the table.

[00:13:34] We'll, dissuade that. Recognize it. You can think of it as prompt discipline or agentic code development, however you think, but there are, there are some contingencies that take exception. So why was it important? Well, the better that you describe what you'd like to receive a large language model, um, the better the outcome. And so in many cases when you hear, I tried it, it didn't work, it came back with some erroneous information and such can really be solved by

[00:14:04] just being more disciplined in the way that you create prompts. Mike?

[00:14:08] mike-richardson_1_08-22-2025_100238-1: Whether you are using chat, GPT or copilot or Claude, what, whichever of these large language models you are using. The quality, you get out what you put in kind of thing, right. We've always said that Right. Put rubbish in. you get rubbish out, right? Yeah,

[00:14:24] ryan-niemann_1_08-22-2025_120238: well, I would tell you that that's a good way to start. 'cause in many cases, I would say a vast majority,

[00:14:30] um, you just put in a prompt and you get something back. And that may be good enough.

[00:14:34] mike-richardson_1_08-22-2025_100238-1: Yeah.

[00:14:34] ryan-niemann_1_08-22-2025_120238: am, uh, I don't

[00:14:35] tom-adams_1_08-22-2025_130237: Yeah.

[00:14:35] ryan-niemann_1_08-22-2025_120238: at my background. If I wanna get a, a new lamp with a glass base, I can put that in.

[00:14:40] It'll probably gimme back some, some good, good, good examples. That's just without doing any kind of, of prompt discipline. I, we've talked about, and I think we touched on this in our last episode, there's something like. as a I will. You will. And so that would be evolving that into act as an interior designer.

[00:14:59] I will give you some options and constraints on my budget and such, and you'll provide me

[00:15:04] some options of different lamps I could buy.

[00:15:07] mike-richardson_1_08-22-2025_100238-1: Just give us that one more time, those three elements.

[00:15:10] ryan-niemann_1_08-22-2025_120238: Yep. Act as a

[00:15:12] I will. You will.

[00:15:14] mike-richardson_1_08-22-2025_100238-1: Nice. So that's step one. Everybody, if you wanna, if you wanna do better prompting, that's a nice little easy to, easy to remember structure. Perfect.

[00:15:23] We'll come back, Ryan, 'cause I know.

[00:15:25] you've got a lot more. But Tom, any any quick reactions and Mark any quick reactions so far?

[00:15:29] tom-adams_1_08-22-2025_130237: yeah. The, the, the thing for me is, uh, and when, when Ryan talks, LLM, there's, there's two directions. One is conversational. I, the thing that's been helpful to me is there's the conversational type of chat. And then there's also what, what is considered product prompting, right? Which is creating agents that respond to things.

[00:15:49] So, uh, my sense is this, this first thing

[00:15:53] is, is more conversational that,

[00:15:55] mike-richardson_1_08-22-2025_100238-1: Right. Okay.

[00:15:57] tom-adams_1_08-22-2025_130237: when you're in conversation, act as if, uh, I'll do, you do kind of stuff That, that, would you agree with that, Ryan?

[00:16:04] ryan-niemann_1_08-22-2025_120238: Yeah, that's a good delineation. I think kind of as Yeah. the progression,

[00:16:08] tom-adams_1_08-22-2025_130237: Yeah.

[00:16:08] ryan-niemann_1_08-22-2025_120238: that they build upon each other

[00:16:10] tom-adams_1_08-22-2025_130237: Yes.

[00:16:10] mike-richardson_1_08-22-2025_100238-1: Okay, mark, how we doing? you are you learning something already? so you keeping up?

[00:16:14] mark-redgrave-audio: look, I, why I think this is so interesting. The first, the first thing is like, um, if you're sat there going, oh, hold on. It's already like feeling like, like just query. It, it doesn't matter. Right? Because, because my, my first impression of this stuff was, was you get such impressive results back. What, what, what, what Ryan's about to dive into is how to make that stuff even better.

[00:16:34] But my first thing is don't hold back. Like just get into it. But why I think this is such a great conversation is this is about teaching, like from a CEO's point of view, teaching your business to think clearly. If you can't describe what you want, maybe you don't have the clarity you need, which I think is super interesting as an area, Right. So it's like, it's almost like the prompting your capability to prompt, becomes a proxy for do you understand what you want? And I, I think that's quite an interesting concept.

[00:17:06] mike-richardson_1_08-22-2025_100238-1: Alright, back to you, Ryan. Keep going.

[00:17:08] ryan-niemann_1_08-22-2025_120238: that's a, it's a good thing to remember Mark. So I would say so then progressing out of as a I will. You will. Um, we talked last week about using a prompt that actually asks it to improve your prompt, like super basic. I think it's interesting that there's a number of tools out there. Some of 'em you have to pay for to improve a prompt, but in this case, whatever. Large language model you're working with. If you provide a prompt and then say, would you improve my prompt before proceeding? It will,

[00:17:37] mark-redgrave-audio: Yeah. Yeah.

[00:17:38] ryan-niemann_1_08-22-2025_120238: say, this is a better prompt to start using. Uh, I also introduced the, the thought of, and then ask me three questions before you proceed, or five questions, or as many questions as you need to get clarity. can really just start to improve the results you get back. And then lastly, you can use something like. your first response against these things that I find important and then provide me yet

[00:18:04] mark-redgrave-audio: Yeah.

[00:18:05] ryan-niemann_1_08-22-2025_120238: and it continues to evolve. The response to getting really

[00:18:09] specific on what you're looking for in a response.

[00:18:13] mike-richardson_1_08-22-2025_100238-1: Oh, nice. Yeah. Tom, any, any additional

[00:18:16] thoughts around all of that?

[00:18:17] tom-adams_1_08-22-2025_130237: Yeah, and I and I'm pulling off what Mark already said. I feel like this is the thing that we learn when we're, we're working with teams, like when we're working with staff members, which is what Ryan just said, is. Often how we have to converse with a person that we're asking to do a specific thing, which is we have to be really specific what we want.

[00:18:37] We gotta say, I'm gonna give you this, you're gonna gimme that. Um, and then can you make sure that you sort of bring back to me what you think I just told you? ask me any questions. You like that, that's what's so profound about this is. teaching a machine that's been taught on the way a human tends to think. Um, and it's not quite as good, but like that's the same process. And to

[00:19:01] mike-richardson_1_08-22-2025_100238-1: Yeah.

[00:19:02] tom-adams_1_08-22-2025_130237: what, what I do when I'm working with it is I keep asking things like as if I were asking a human and I, I give them sort of personas so that I can talk and think like that.

[00:19:13] mark-redgrave-audio: Tom, I, had a conversation with a, not a CEO, but a senior leader yesterday. and one of the, things we talked about was like the need for leaders to seek clarity. It's, isn't it is, is that's, that's what you do. Right? So I think this is really, like this is perfectly aligned and I, but I've got a question for Ryan.

[00:19:29] Very specific thing which is like, when as we're, evolving the prompts, right, and I'm, and I'm giving, in this case chat, GBT, more clarity, like it always agrees with me. So how do we fight that So, so for example, I'll give you an example, right? So, so I go, oh, it fee, so, so they might come back and I say, it feels like we are focusing too much on this and not enough on that, because they, this thing that I think we should focus on is important to CEOs and and, And often it comes back and goes, oh Yeah.

[00:20:00] mark, you are right. do this. And I'm like hold on. Don't agree. Okay, so what do we do about that?

[00:20:06] ryan-niemann_1_08-22-2025_120238: Actually, that's a good point. I've used different, uh, language models and I do believe it was, llama that

[00:20:14] came back and like at the end said, how's your day?

[00:20:21] tom-adams_1_08-22-2025_130237: Right, right.

[00:20:21] ryan-niemann_1_08-22-2025_120238: Okay, so without getting technical, there are things that you can control when creating a prompt there's things that you don't control. And the default settings are typically things like be a helpful assistant

[00:20:37] and be friendly in responses and

[00:20:39] mike-richardson_1_08-22-2025_100238-1: when you say, Ryan, when you say, when you say default settings, you mean in instance in chat, GPT. If you go into the. settings area.

[00:20:47] ryan-niemann_1_08-22-2025_120238: No, no, no, no.

[00:20:48] This is deeper. This is

[00:20:49] not, it's not, you know, there may be other, uh,

[00:20:52] let me clarify. There may be some language models that you can toggle and,

[00:20:56] mike-richardson_1_08-22-2025_100238-1: Oh, okay.

[00:20:57] ryan-niemann_1_08-22-2025_120238: and don't get me wrong, early on, those things were really kind of available to users,

[00:21:01] but now they start with a default setting.

[00:21:03] You can't control that in

[00:21:05] mike-richardson_1_08-22-2025_100238-1: Oh, I see.

[00:21:06] ryan-niemann_1_08-22-2025_120238: So, so then you can say like, when instead of a helpful assistant, I need you to be

[00:21:11] a engineer with a PhD in electrical engineering. Well,

[00:21:16] mark-redgrave-audio: Right,

[00:21:17] ryan-niemann_1_08-22-2025_120238: different than

[00:21:17] being a helpful assistant,

[00:21:18] tom-adams_1_08-22-2025_130237: Yeah.

[00:21:19] ryan-niemann_1_08-22-2025_120238: And I, need you to be critical of my ideas and challenge my

[00:21:24] perceptions like that. That's different

[00:21:26] than being agreeable and going with the

[00:21:28] mike-richardson_1_08-22-2025_100238-1: Okay. Okay.

[00:21:28] mark-redgrave-audio: but.

[00:21:29] ryan-niemann_1_08-22-2025_120238: Mark, I'm gonna

[00:21:29] go one deeper for you that I found this weekend, and the good news is you can actually ask any large language model to help explore this topic. So don't get confused if you're like, oh, this kind of sounds interesting. I'm gonna learn more. Just prompt. LLM to give you some information about it, but it's a sixth component from Prompt Framework.

[00:21:49] And the reason

[00:21:49] why I found this out is one of our listeners said that they had got some value in being better at prompting and the like.

[00:21:56] And so I did some research and with chat,

[00:22:00] mike-richardson_1_08-22-2025_100238-1: We have a listener.

[00:22:01] ryan-niemann_1_08-22-2025_120238: well, I say

[00:22:02] that it was, it was, it was somebody that had responded, I think to a LinkedIn post that was your post, and they said that was really valuable.

[00:22:08] So I

[00:22:08] mark-redgrave-audio: I, I told my wife to stay off the internet, Ryan, so I apologize.

[00:22:15] mike-richardson_1_08-22-2025_100238-1: Sorry, Ryan. Keep going. Keep going.

[00:22:17] ryan-niemann_1_08-22-2025_120238: And the, uh, so here's, here's the construct. I found this out because of Chatt PT five release. if you go to the OpenAI platform, which is the platform

[00:22:28] below chat pt, they have something called a playground. They actually have a prompt builder. can, it's, it's, included. If you have a subscription, you can explore it there. and the, the six different components are role, context, reasoning, format and success, and able to provide the LLM some amount of information in those six categories,

[00:22:53] mark-redgrave-audio: Okay, give your name again, Ryan. Gimme them again. Roll.

[00:22:56] ryan-niemann_1_08-22-2025_120238: you bet, role, task, context, reasoning. Format and success, and I'll give a qualifier. I did some research this morning. There is no one universal six components. It doesn't seem to have a, a distinct origin story, but it is

[00:23:13] very well recognized that these things, can, help improve your prompt. And there's some variations uniquely. One of them is actually tone. For instance, the tone could be more professional versus more casual

[00:23:25] mark-redgrave-audio: Yeah.

[00:23:26] ryan-niemann_1_08-22-2025_120238: more

[00:23:26] technical, versus more, uh, you know,

[00:23:29] uh, theoretical.

[00:23:31] mark-redgrave-audio: Love that.

[00:23:33] mike-richardson_1_08-22-2025_100238-1: And are you able to sort of run us through an example, Brian, of how you would, how you would, uh, the di those six, how you. would construct those six dimensions for an example? Prompt?

[00:23:44] ryan-niemann_1_08-22-2025_120238: Yeah, you bet. So I would start by, describing what you want to accomplish. I

[00:23:49] would dive in on a particular role, maybe add an industry,

[00:23:54] add a particular, professional field,

[00:23:57] mike-richardson_1_08-22-2025_100238-1: Mm-hmm.

[00:23:58] ryan-niemann_1_08-22-2025_120238: add a particular discipline, uh, maybe years of experience. Uh, in that you could also say, you know, requires technical knowledge of HIPAA compliance, or, uh, requires technical knowledge of a particular

[00:24:14] CRM system at, at your work, right?

[00:24:17] mike-richardson_1_08-22-2025_100238-1: that was the first element, which was role? correct, Okay.

[00:24:21] ryan-niemann_1_08-22-2025_120238: moving on to task now. The task and, and how AI approaches it is kind of interesting and I'll make this very simple. If you ask it to do something, uh, and it comes back, it it, you're giving it instructions without examples. Uh, another approach is that you provide

[00:24:39] some examples of what you want in the response. you know, if I were asked Mark to go get me a piece of paper, really there's no instructions on what I want back. But if I said, would you go, gimme a piece of paper that's lined and also is four by six starts to get a better contextual understanding of what I want back. And the last example is what's called chain of thought, right?

[00:25:07] So what does that look like? Well, for instance, what if I asked Mark to work through a series of questions? And through that process I really got a better understanding of what I wanted back, which could be something like a. Hexagonal graph paper, like I didn't even think of the fact that I needed a hexagonal graph paper.

[00:25:28] Maybe I didn't even know what that was, but that's exactly what I needed. And through that chain of thought, mark was able to elicit a better understanding of what I wanted back.

[00:25:36] mark-redgrave-audio: Is it Ryan is um, 'cause I, uh, this is how I emotionally react to writing prompts, right? I want like, I'm normally pretty excited and I sit down and I start writing and I wanna press enter. Like, I, I totally wanna press enter, but like. I force myself to like know more information more. So one thing I, I, I wonder is, And this is just from my personal experience, is like

[00:26:05] you should be prepared to sit and take time to write a

[00:26:12] great prompt, right?

[00:26:14] Because I just find myself rushing often. Yeah. So what's your views on that Because I know you can refine prompts, but it's like? how much. Yeah, because, 'cause I I I said shit that this, this prompt is 150 words long. And I'm like I wrestle with that Does that make sense or am I maybe I'm talking

[00:26:29] gibberish, but.

[00:26:30] mike-richardson_1_08-22-2025_100238-1: No, no. Go through it, Ryan.

[00:26:32] ryan-niemann_1_08-22-2025_120238: Yeah, my, my first impression is, no, not all, all, all prompts require that level of discipline. I'm sure we could come up with some sort of matrix that like, you know, impact, uh, low complexity, don't spend a lot of

[00:26:46] tom-adams_1_08-22-2025_130237: Right.

[00:26:46] ryan-niemann_1_08-22-2025_120238: prompts, high impact, high complexity. Spend a lot of time on your prompt because those individuals that are in the middle that are like, ah, I just didn't get what I

[00:26:57] wanted.

[00:26:57] mark-redgrave-audio: Yeah.

[00:26:58] ryan-niemann_1_08-22-2025_120238: Quite likely didn't spend that level of discipline and it was an important thing or something that they do quite frequently, and they needed the output in a specific way. And through some good prompt discipline, you can achieve

[00:27:10] some very interesting outcomes that save you time.

[00:27:14] tom-adams_1_08-22-2025_130237: this is where the

[00:27:14] distinction between conversational prompting and. Product prompting starts to become really critical because if you're building a agents, the prompt becomes seriously important. because essentially what you're doing with an agent is you're, you are amplifying the recurring nature and you get leverage from a prompt that's actually well constructed.

[00:27:40] But if we. If we don't learn how to do effective prompts on that sort of conversational stuff, once you get over into age agentic stuff, you don't actually then

[00:27:51] produce the kind of prompts that, that you can scale very well with.

[00:27:54] mike-richardson_1_08-22-2025_100238-1: Okay, so doing, doing conversational prompts in chat, GPT, copilot, Claude, et cetera. That's, that's good. That's good training. That's putting the training wheels on. We'll reap the benefits of that at the next level if and when we go to agentic kind of stuff. Right. Okay. And Ryan, let me ask you this. So is there any difference between if I've got that, high impact, high complexity thing I'm trying to do, is there any difference between me taking 10 minutes to work through those six elements as part of my first prompt versus doing element number one as my first prompt?

[00:28:28] See what I get? Then going back in and doing element number two as my second prompt and doing it iteratively. In your experience, is there any big difference between those two approaches?

[00:28:38] ryan-niemann_1_08-22-2025_120238: There is in the, in the sense that. This was explained to me, and it makes a lot of sense and that large language models are trained to quickly provide a response. and so if I give a project

[00:28:50] detail and then I wait to give it role and, uh,

[00:28:54] mike-richardson_1_08-22-2025_100238-1: right.

[00:28:54] ryan-niemann_1_08-22-2025_120238: and content, it's gonna go off and come back with something.

[00:28:57] But that may be okay. You

[00:28:59] know, you might wanna take a more iterative approach. Um,

[00:29:03] mike-richardson_1_08-22-2025_100238-1: Okay.

[00:29:03] ryan-niemann_1_08-22-2025_120238: you might find yourself starting over or

[00:29:05] like then saying, now I also want to add in

[00:29:09] additional examples of the format that I wanna get it back in. Or I want to add some additional details around the role skills

[00:29:17] that I think

[00:29:17] are important in this project that I'm trying to complete.

[00:29:20] mike-richardson_1_08-22-2025_100238-1: But it sounds like the more that. we want it to think, the more we need to think about first about the

[00:29:29] prompt.

[00:29:30] ryan-niemann_1_08-22-2025_120238: Yeah, you bring up a good point. What comes to mind is, you know,

[00:29:32] if you've got

[00:29:33] some good prompts, you know. Keep them around, like

[00:29:36] mike-richardson_1_08-22-2025_100238-1: Right?

[00:29:36] ryan-niemann_1_08-22-2025_120238: into a, a notepad and then the good news is to Tom's point, you don't need to know code. You can change the wording and

[00:29:44] modify, some of the aspects and

[00:29:47] mike-richardson_1_08-22-2025_100238-1: Yeah.

[00:29:48] ryan-niemann_1_08-22-2025_120238: comes back that's differently.

[00:29:49] mike-richardson_1_08-22-2025_100238-1: Awesome.

[00:29:49] ryan-niemann_1_08-22-2025_120238: Also, do this, try different LLMs, like try a prompt. In chat, GBT. Then try that same prompt in Claude. Then try that same prompt in perplexity, for instance, and see the different styles that you get back. I mentioned llama. I thought that was funny that, you know, it asked me to have a great day. It's like, are you know, are you having a great day?

[00:30:13] Like,

[00:30:13] you know

[00:30:14] that,

[00:30:14] mark-redgrave-audio: Ryan, you never ask alarm or anything. you know. that. What are you gonna get outta alarm?

[00:30:18] ryan-niemann_1_08-22-2025_120238: Yeah, that's a good point.

[00:30:19] tom-adams_1_08-22-2025_130237: So uh, mark.

[00:30:21] ryan-niemann_1_08-22-2025_120238: me. I was actually talking to a llama.

[00:30:23] tom-adams_1_08-22-2025_130237: Yeah.

[00:30:24] mark-redgrave-audio: Anything knows they dunno. Nothing.

[00:30:28] tom-adams_1_08-22-2025_130237: Mark,

[00:30:28] mike-richardson_1_08-22-2025_100238-1: Okay.

[00:30:29] tom-adams_1_08-22-2025_130237: gonna give you a way to take Ryan's complexity, and state it all 'cause you're in a rush, which is talk to it and

[00:30:38] I use whisper tools, so I actually talk

[00:30:42] Instead of typing and I talk through it and I, I don't care how

[00:30:46] long it takes because my whisper tool that I talk to actually

[00:30:51] cleans it up and puts it into the prompt window. And so I don't have to type my way through a process. I think my way. 'cause you talk like, your brain functions really well when you're talking as much as I've noticed and sometimes it's way ea, I find it way easier. I just

[00:31:06] mike-richardson_1_08-22-2025_100238-1: Right.

[00:31:07] tom-adams_1_08-22-2025_130237: prompts all

[00:31:07] the

[00:31:08] time.

[00:31:08] mark-redgrave-audio: So what's Whisper? What's whisper Tom? What is that exactly.

[00:31:11] tom-adams_1_08-22-2025_130237: there's, tools Right. now that are basically apps and you press a, a hot key on your keyboard and it just records what you're saying,

[00:31:18] mark-redgrave-audio: Okay,

[00:31:19] tom-adams_1_08-22-2025_130237: it, it,

[00:31:19] sends it to like an LLM to clean it up and put it in a particular format, and then it drops it into the whatever box. And so I never

[00:31:27] type, I literally

[00:31:29] spend all my day just talking.

[00:31:31] mark-redgrave-audio: love that.

[00:31:32] ryan-niemann_1_08-22-2025_120238: this is a brilliant distinction 'cause I don't do that.

[00:31:35] I, I really do not.

[00:31:37] Very rarely outside

[00:31:39] mike-richardson_1_08-22-2025_100238-1: right. Yeah.

[00:31:40] ryan-niemann_1_08-22-2025_120238: or Alexa and things like that, I really don't find myself spending time, uh, doing

[00:31:45] it

[00:31:46] that

[00:31:46] mark-redgrave-audio: as, as somebody, as somebody who recalls a lot of voice notes, Tom, like, yes. So I'm like, I'm like when I'm in the.

[00:31:52] tom-adams_1_08-22-2025_130237: All day

[00:31:52] long.

[00:31:53] mark-redgrave-audio: Imagine if I'm in the car and I can start to just voice record things that. get turned into prompts that then

[00:31:59] get triggered while I'm in the car. I'm like, okay, now we're talking.

[00:32:02] Right.

[00:32:02] mike-richardson_1_08-22-2025_100238-1: yeah. So, Ryan, uh, correct me if I'm wrong, but I think you took a few minutes over the

[00:32:07] weekend to actually put these six steps into a URL, didn't

[00:32:12] you?

[00:32:13] ryan-niemann_1_08-22-2025_120238: I did. Yep.

[00:32:14] our listeners can find [e-mail].

[00:32:19] mike-richardson_1_08-22-2025_100238-1: Prompt Works Pro. Everybody go over there.

[00:32:23] tom-adams_1_08-22-2025_130237: I've been trying it and it's wicked cool. Wicked cool.

[00:32:27] ryan-niemann_1_08-22-2025_120238: Just, uh, you know, uh, Mike, I got your feedback and I added a, a chat bot agent to it as well. So there's a, a chat bot to help, uh, you and others, uh, be successful in creating a sixth component prompt.

[00:32:42] mike-richardson_1_08-22-2025_100238-1: Prompt works pro everybody. We'll put that?

[00:32:44] in the show notes as well. Go check it out. Okay, everybody. We're gonna start landing this plane with the final element in the sandwich here, which is what is something that would be doable this week or these two weeks before our next episode for our listeners to head out and do, Mark, what would you suggest for our listeners to bite off something that's do doable and digestible?

[00:33:08] mark-redgrave-audio: So, I would say if you are a, so this is a thought that I'd like to lead, leave CEOs with to take into

[00:33:16] this week, which is like,

[00:33:18] you know, prompting is gonna be table stakes.

[00:33:22] everyone is going to need to know, how to prompt, right? So I would say think of that like a, core business skill.

[00:33:31] So if you're a CEO go into this week and, and start having conversations with your team saying, right. We need to build capability. You have those conversations all the time. Right? there's a capability we need to build here, which is called like, I wouldn't call it prompt engineering. It sounds too scary, but we have to make

[00:33:47] sure all of our team know how to

[00:33:49] prompt effectively.

[00:33:51] So start planning that yesterday. I, I think that's because I think that's a really, like, that is where this is headed really quick. Right. And, and why to, just to loop back to the comment I made at the start of the show, it's like. If your team can't have the clarity to prompt properly, they probably don't understand well enough what you're trying to do.

[00:34:13] That's a strategy gap. So it's like, it's so, that's so interesting, right? Where these two things glide, so, so that would be my thing for this week

[00:34:20] Go and have a conversation with your team about like, how are we building prompting as

[00:34:24] a core capability of everyone in the company.

[00:34:26] mike-richardson_1_08-22-2025_100238-1: Yeah, and I like what I like what Ryan said earlier as well. Maybe he was talking mostly about individual listeners on the call, which is, hey, if you discover prompts or the six structure of prompts that work, copy and paste those to accumulate that inventory into OneNote or a Word document or somewhere that you can get at quickly and easily so you're not having to reinvent the

[00:34:48] wheel

[00:34:49] mark-redgrave-audio: Yeah, it's an, it's an asset. It's an asset, a co. You need to, as a CEO, you need to think about the, your, your prompts is becoming a corporate asset.

[00:34:58] mike-richardson_1_08-22-2025_100238-1: yeah. Beautiful. Great. Uh, Tom, what would you recommend to people as a doable thing? They can bite off over the next couple of weeks.

[00:35:05] tom-adams_1_08-22-2025_130237: Well, like I said, I'm practical and I'm gonna jump out of the prompt side of it. 'cause I think this is

[00:35:09] really prac. I, I just love small problems that recur regularly and if you can figure out how you wanna solve those, you now have the opportunity to create a prompt to create a result in some kind of agentic format.

[00:35:25] But so often what I

[00:35:26] hear is people tend to

[00:35:27] think in terms of big stuff instead of

[00:35:30] mike-richardson_1_08-22-2025_100238-1: Yeah.

[00:35:30] tom-adams_1_08-22-2025_130237: things. Like

[00:35:31] what's

[00:35:31] the

[00:35:31] mark-redgrave-audio: Yeah.

[00:35:32] tom-adams_1_08-22-2025_130237: problem like. Uh, I had a conversation with somebody this week and they said, my, I, I'm so frustrated by my email. Well, what's the frustration? I've gotta tag everything. Well, I go, well, that's, that's a problem.

[00:35:43] That's solvable, but it's, let's not, let's not think about AI as this whole thing that comes in the business. Let's look at a tiny problem that you can

[00:35:51] actually solve. That recurs on a regular basis, and just know what you want from that, and then you can create a prompt and said, well be clear about it.

[00:36:00] My clarity is I want this done right? You're gonna do this and I

[00:36:04] want my emails tagged every five minutes. Something

[00:36:08] mike-richardson_1_08-22-2025_100238-1: And, and as a, as a small example of that what I've done for this podcast is I've built myself a little project inside of chat GPT with a number of different chat channels in it where I'm able to go in and and ask chat GPT, Hey, what's the new news that's, out there? I want you to go out and bring me back a digest.

[00:36:27] And I just went in there this morning and I said, please refresh the digest with the very latest. Boom, there it is.

[00:36:34] and then there's another channel in which I've said, look, I, my angle for these podcasts is this. Help me brainstorm. You know, I've been doing this, this, this, this, and this. Help me brainstorm what my contribution can be today.

[00:36:46] So I can just go back into those channels and I can almost just sort of say, please refresh and gimme the very latest. And it's a two. It's a, it's a 22nd exercise.

[00:36:57] ryan-niemann_1_08-22-2025_120238: you just reminded me of something, uh, and, and leaning in on the, you know, glass of water for every prompt. Don't ask it, please. And it's interesting. It's interesting for two reasons. One, I, I've actually found that

[00:37:13] some responses will be better if you challenge it and you

[00:37:17] mike-richardson_1_08-22-2025_100238-1: Yeah.

[00:37:17] ryan-niemann_1_08-22-2025_120238: isn't this the best you can do? Or like, or think harder about this? And I did. There's a little nuance to the GPT five release where she said, well, you can access this to think harder about a topic, because that's the default setting if you're a paid subscriber.

[00:37:34] If you're not, you can trigger that by saying,

[00:37:36] think harder about this. There's a tip

[00:37:39] for

[00:37:39] everybody.

[00:37:39] mike-richardson_1_08-22-2025_100238-1: And over and above all of that. Ryan, what would you recommend to our listeners as a doable thing? They can bite off this. week or next week.

[00:37:47] ryan-niemann_1_08-22-2025_120238: You know, I had a, a conversation with a leader and I encouraged. individual to create a, a council or a small team to get started. Um, I, I think a, a lot of individuals are

[00:37:59] challenged with, how do I get started? I don't know everything. And the good news is no one knows everything

[00:38:04] tom-adams_1_08-22-2025_130237: Yeah.

[00:38:05] ryan-niemann_1_08-22-2025_120238: and surrounding yourself with some bright talent in your organization or in, in your, in your firm, your business can, or in education in this particular context, you know, get some of those students. come and be part of a council and start to outline how do we take next steps? How do we, how do we make this meaningful and impactful? Really with an effort to grow to something that really sig saves significant time or is really impactful on people's lives or actually is able to demonstrate significant ROI, any one of those objectives?

[00:38:43] How do we take steps and

[00:38:44] really continue to grow and learn and make it meaningful?

[00:38:47] mike-richardson_1_08-22-2025_100238-1: Yeah. Yeah. Thank you. And for me, everybody, I'm, I'm, I'm gonna keep playing with my, my Claude note pin here. And, whether you are using, you know, one of these AI recorders or you're using the Zoom Z Zoom AI companion or, or the equivalent in teams or copilot or whatever, I urge you to, to just get in the habit of using these AI companions for every meeting that you have. and what I love about this sort of note pin is that, you know, I can leave it on all the time. It's got the battery life for that. I can be sending myself little verbal notes, Hey, don't forget this. Uh, here's an action item on that. And then at the end of the day, when it's uploaded, I can just interrogate it for, Hey, what are, what are the action items that I identified today?

[00:39:34] What were the things I needed to not forget today? And I just find it's, it's helping my productivity a lot. So I'm, I'm watching a lot of YouTube videos now about people who've been using these things for a while, and they've really started to figure out. How to best apply it and deploy it into their life.

[00:39:50] And they, you know, they just talk about how it's, it's changing their whole productivity and getting a lot of low value, high, high effort workload off their plate. And, uh, that's, uh, that's what I'm gonna be experimenting with, uh, this time around. anything else for the good of the order? Everybody did, uh, did we get to everything?

[00:40:07] Mark, Ryan, Tom, that you wanted to share today? Anything else that you want to just throw out there before we leave?

[00:40:13] tom-adams_1_08-22-2025_130237: I think we covered lots. I think that's

[00:40:14] mike-richardson_1_08-22-2025_100238-1: Awesome. Awesome. Well, there you have it everybody. Another episode of the Prompt and Circumstance podcast, in the can as it were. go check us out at prompt and circumstance podcast.com and we look forward to having you with us next time.

[00:40:29]