Guys. Welcome to the podcast. My guest today is Nandanae Sabharwal, and she is a field medical director with the migraine and women's health group at Pfizer. And we talk about the importance of using artificial intelligence in everyday MSL life. We call this the augmented MSL. So it's great conversations. Very, very important in this day and age.
Don't forget there is still time to sign up for the Medical Affairs Excellence Summit, MAACS, and that's October 22nd 23rd in Atlantic City, New Jersey. To for more information, just go to the Internet and type in Max Summit 2024. I look forward to seeing you there. It's gonna be a great event. There's still time, so get your registration in. Welcome to MSL talk with Tom Caravella, a podcast specifically designed for MSLs and all things field medical. Hey, Nandanae. Welcome to the podcast.
How are you? Hi, Tom. Thank you for having me. Yeah. I'm excited. So, guys, I mean, I guess I should go back to how we know each other. So Nandan and I met through Charlie Cook, who's been on this podcast many times. He had an open position. We had met, and the rest is history. She's been on now are you still on Charlie's team? No. I ended up on Charlie's like a different team. Oh, okay. Alright.
So but still, that's that's kind of the history, and we just have kept in touch, met actually saw her at Mass East. That's why it's important for you guys to go to conferences. It's very good to keep up relationships and networking and see people and that good stuff. And she had this amazing idea to talk about the augmented MSL and how to use AI in your normal everyday life as an MSL. So this is gonna be an amazing conversation, but I'm gonna let her do an introduction to start.
Tom, this is exciting. Dee, I wanna also thank all of my predecessors who have done a podcast with you, Tom. You guys have made me smile, chuckle. I would laugh as I would listen to you on my long drives of South Jersey for my HCP meetings. So I certainly, you know, felt so supported by the MSL community that I wanted to share what I've learned on my journey as an MSL who is fortunate enough to be able to use some generative AI or Gen AI for short.
So just using some of these tools in my daily life. The I'm currently a field medical director on the migraine and women's health team adviser. I'm in Jersey. The I've spent about 5 years at big pharma as an MSL. I am a clinical pharmacist by training. I was trained in the South Bronx. I have spent 5 years going across every settings. So as a disclaimer, I have zero knowledge in coding.
I have never taken a c plus plus class, but what I do have is proficiency in prompt engineering, which I was able to self taught myself. Prompt. You are the prompt engineer. I love that. That's, like, such a modern term. Like, a prompt that who would like, 3 year a year ago, someone would say, oh, do you know what a prompt engineer is? I'd be like, I don't know. Someone who's on time? Like, someone who's always prompt? Like, I wouldn't know what that is, but that's actually a thing.
Well, we'll get into it. Actually, are you ready to get into it? Should we jump in? Yeah. We can we can jump in. Because I wanna start with you. I wanna start with your journey, how you jumped into AI, and what that learning process was like. Because I want people that haven't really jumped in yet. I want them to hear from someone who has. So my advice was become an early adopt become an early adopter of AI. Mhmm. And so when, you know, this this became really popular a year ago.
And so I I downloaded the Chat GPT app on my phone. And so, you know, I slow it was a slow start. Yeah. And then so, for example, I would be in Whole Foods and I'm, like, looking at all the I'm like, what is the difference between a tomato puree and a strained tomato? I have no idea. So I go and ask ShajiPT and it gives me exactly what I need to know. So it was just like, as a busy working mom, I just need answers a lot faster. I just don't have the time to, like, scroll through web pages.
And then so just seeing how it benefits my personal life, I was convinced it was going to help me in my in my job as well too. So you're even using it from, like, in from not, I mean, just a personal standpoint, even, like, recipes. So, like, if you have this, like, idea to make something, you throw it into chat GBT? Oh, yeah. It will give you like it will give you recipes. You know, you can ask it for like this. These are ingredients. I have my my kitchen.
You can take pictures of your ingredients and it's going to create what you already have. So that's that's the cool part of it. So as I and then so I would challenge myself instead of like me trying to go to Google, I would consciously stop myself and say, 'Okay, let me ask chat GBT and see what it says.' And then so slowly started using it a lot more.
So a couple of times a week, and then I started using it daily and then I almost became annoyed at Googling because I'm like, I'm just scrolling, scrolling, scrolling. I can't find the answer. Google don't Google has has redeemed itself, though, because now it has, Gemini by Bard and it will automatically, you know, find those answers for you right away. So it has redeemed itself. So now I Google a lot more now.
Yeah. But now do you is that your would you say that's your advice to MSLs that, like, are gonna get started? Like, if they're looking to, say, upscale into this AI world, is that the way to start? Is it like, hey. Just play around with it with just everyday kind of stuff and practice prompts and see what comes up. Is that what you're kinda saying? Yeah. That's a great way to start, using things in your everyday life.
You know, you can arrange something like, you know, something that's a very simple question, to something that's a little bit more complex. And that's where you kinda not have to know with prompting with prompting. Prompting, all it is, it's just a communication style. Just like MSLs, we are have become we learned to be effective, like, communicators. And that's because we've learned how to adjust a different type of style. And it's the same thing. These are something like ChatPQPT.
It's a language model. And you're just kind of figuring out, okay, how is this system, how do I communicate with this system? And then, so the prompt itself, if you're looking at something, you're solving for something that's more complex, you're gonna have to create a more a more sophisticated prompt. In other words, so in its most basic form, you are giving context and you're giving task into chat gpt.
And then so the more context you give it, the better task you give it, it's gonna give you a better answer. Yeah. Well and I could I'll add to that because I and, again, I don't know what a prompt was. They're like, that's the question that you asked, you know, chat gbt. That's the prompt. That's the question. And in the beginning, I would just ask a question. And then a lot of times it wouldn't be specific enough or detailed enough where the answer would would actually make sense.
So what I started to do was give more detail, be more descriptive, and really command Chat GPT, but also educate not educate, but inform of what I was doing, what I wanted, what I think I want to add to whatever the situation is.
And then give it enough parameters where it's gonna take let's just say you wanna write an article, and you have an idea, and you have some points, but you don't have the whole thing fully baked, and but you have enough to get started, but you're like, I just don't know where to go with this. That's the perfect application because you can give chat GPT all those details and then let it do it for you. Is that what you found as well? Yep. And you'll learn it by trial and error.
So you can ask it for you know, if you give it a question, your prompt, and then you don't get the desired result, you can ask it, you know, can you please make suggestions on how for me to improve on this prompt? You know, what other questions would you consider asking in this scenario? And then it will give you those it will give you some of those responses. And then so that you can use it as a way to help you improve your prompts. I, you know, I never knew that.
I I didn't realize that you could actually Yeah. Ask it to double check your prompt and, like, improve your pump. That's awesome. Okay. So if we ended this podcast right now, I definitely learned something, but we're not ending it. Let's talk about let's look at, like, the com we're talking about the individual, but let's talk about the company for a second. Like, how important is company buy in, and is there would you say now you're with a big company.
You're with Pfizer, so I think it varies from company to company. But, like, I'm curious as to how important company buy in is and if companies are really directing MSLs thoroughly on this, or does most of it have to fall on the MSL at this point? See, I'll share my journey. I mean, of course, like, these are my experiences, and these are things that I, you know, I've kind of learned on my own. And it and it could be different from company to the company. Having c suite buying is critical.
Having so a company who uses gener Gen AI tools is gonna have the competitive edge over companies who don't use it. And then so for the C suites, they have to they have to believe in this. And it's certainly a big investment upfront for them. And then so it's kind of thinking, you know, it's very expensive for, you know, purchasing cloud computing services to run things like language large language models.
And so the best companies who are able to scale AI within their company are going to be companies who embed this into their strategy. And they are and they're going to train their employees on how to use it. And then so that's like a big that's a, you know, a big hurdle right there. So if, you know, your company is able to get over that hurdle and you get the opportunity to use it, even then for me, I still had to have buy in for my team.
So these these Jenny I tools, these are things like Microsoft Copilot, something like a long a large language model like Chat GPT. They're gonna cost money. There's a license, like, a cost associated to it. So to get you know, I had to get the support, talk to my manager, what this is, when why do I wanna use it, get their get their support, and I need somebody to give me the money to be able to purchase it. So the I also had to so I had to write up a proposal to my senior team leader.
And in my proposal, I'm telling them, you know, this is the pilot that I wanna do. I wanna, you know, be a test cases, and I want to measure standard approaches. You know, the way of working as an MSL versus using a Gini AI tool in increasing my productivity. And then I'm going to share my learnings, you know, after I finish the pilot. And I'm also going to I'm also going to be teaching my colleagues as well, too.
So, you know, and then once I get the blessing, then I'm able to, like, actually, like, get access to it. So it is yeah. That's, you know, that there still is, you know, buy in that's required from your manager or your seem your senior team leader. That makes sense because you need the tools. But that said, ChatGPT is, like, $20 a month.
So, I mean, even if you had to, like, kinda pay for that out of your pocket or try to get that expense at some point, there there's no real barrier of entry to an extent. Right? But Microsoft Copilot, is that just Microsoft's version of ChatGPT? Indeed. I will say that, you know, when it comes to using these these tools, like, only use it if your your your company explicitly, like, lets you use it and they have a license for it.
Using it, anything that's work related, don't use it on your personal accounts. It's just there's a lot of security and privacy ramifications for it. So try to separate out the 2. Yes. There is. So for Microsoft Copilot, and I think they've just released it just for so for individual subscribers. So if you have access to a Microsoft account, then you can get the premium version of Microsoft Copilot. And in Copilot, since it's powered by OpenAI, it is very similar to ChatGPT.
And then so the nice thing about it is that it's going to speak to it's gonna work across all of the Microsoft platforms, so Word, PowerPoint, Excel. And then so you can work off of those applications. And then even if, even if you need to like, Microsoft Copilot has access to the web itself too. So you can prompt and ask the questions there, and then you can actually have it pull in information into Word, and then you can have it wordsmithed.
You can ask it to convert that over that in content into press PowerPoint presentation. Now it doesn't do it perfectly, but it will cut down on a lot of those clicks and tasks that you have to do. Got you. Awesome. Okay. That's helpful. It set the tone for because I'm really curious to learn about use cases and practical applications.
So talk to me about how you're using it and how MSL should be using it in their job with what types of app in what types of applications, and what are some of the use cases that would make sense? My advice would be to just think about the pain points in your role as an MSL. And then so how can how can Jenny and I help you with this this pain points?
It's not gonna be able to help you with everything, but some of the things that I've been able to use it to do are things like, for example, meeting preparation for a health for a health care provider. Mhmm. And if you, as an MSR, are new to that therapeutic area and to the specialists within that area, it takes a significant amount of prep time to gain foundational knowledge within that therapeutic area. So this is where you can go in and you can ask SHAD GPT, hey. I'm an MSL.
I'm new to this therapeutic area. The these are the specialists that I'm meeting with, for example, a pediatric epileptologist. And, you know, I wanna know what are the most common challenges that this provider would face with diagnosing rare seizure disorders. What are their challenges with managing it? Patient with managing it? What are their what treatment options? And then you can upload, like, for example, your PowerPoint, your PADCARE MSL decks into there.
Say in presenting this content to the pediatric epileptologist. You know, what kind of questions would they have on this content? Or what and then so it'll give you and then give you, like, really good questions. So you can say, okay. Let's role play for a second here. And then you can say, you can tell Taghipt you are the pediatric umpathologist, I'm the MSL, you know, I'm presenting this content to you, we're talking about treatment based guidelines.
You as a provider is giving me rebuttal in terms of my I'm good with my current practice. The current meds work. I don't need anything new. You know, what can you know, how can I incorporate this? Can you help me create a 2 minute conversational script and how we can go about having this conversation? And it'll create a really nice script, and it'll incorporate the content that you give it into it. So as you're going through and preparing, like so that's what it does.
It it really cuts down, like, that training time that you that you need. So it speeds up that learning curve. If you are so that's one of my favorite things to do with it, like role playing. If you are running an advisory board, for example, and you're like, okay, usually the decks are from a clinical trials, and then so you can upload a clinical trials deck. And then you can say, hey, this, you know, can you create I'm running an advisory board.
These are the type of specialists that are gonna be there. You know, based on the uploaded file I gave you, can you create some pre survey questions? Can you create some post survey questions? And then so it's gonna be able to do it for you. So granted, because your company has a license to it, that information will stay within the company. It's not gonna be shared outside of it. And so your all of your data, it's all protected. So that's that's one thing I wanted to add.
If you wanted to, you know, run if you were as you have special projects in your in your role. And then so you're like, okay. You know, how do I, you know, get this started? And then so you can start asking, hey. I have this leader. I have this project. This is what the project is about. And, you know, can you help me create an exercise or create a step by step framework on how to go about executing this project? And then it will give you a, like, you know, a great starting point.
And then from there, you can start asking it more questions along the way too. Sometimes some of the systems will give you suggested prompts as well too, so you can just continue that conversation. If you have, like, post congress summaries and you you know, it's, all of the MSLs are writing out their summaries, and then you have to go over and review everything. You could take all of that, put into TagTBT and say, hey. I'm just coming from conference.
These are all the insights that we've gathered. Can you analyze it, identify trends in it, and then kind of, you know, quantify the data. And then so you can give it more specifics of what you wanna see, and it's gonna be able to do that. So there is a number of things that you can use it for. And then so all it's, you know, just really cutting down the number of, like, tasks that you have to do, in order to get that information.
And I'm sure once you start using it, you you then realize how many other way like, I'm sure, like, as you like you said, as you come up with certain problems or you're faced with a task, you're faced with a a project of some sort, you then need to think about, okay, what in what ways can I use AI? And what whether it's Microsoft Copilot or ChatChipti, whatever it is that your company is giving you.
How can I use that to make this job easier or to take off some of the responsibility of me having to actually do? It's almost like you're you're trying to look for an it's like having an assistant. Somebody who's giving you who's helping you with at least part of the task. Right?
Yep. Yep. And it's so it's not that's the one thing that I've learned, and I've heard it from others too, is that, you know, if you something like a chat gbt, it's not gonna help you, create something from scratch from, like, 0 to 100%. It's going to help you. You, you itself have to come up with the con with the concept itself. So you have to do the first 20% of the work. And then what ChatTBT will help you do is complete like 20 to 80%.
And then the last part of it, you have to put the finishing touches in into it. And then so, you know, if you clearly, if you have an idea, you have to know what you want. So, you know, in there in other words, your input into it. And then you kinda have to have an idea of, you know, what you want, or what the end result or end point is. And then the system is gonna help you in between. And that's that's kind of how you'll be able to use it effectively. Okay. So let's go back.
So you get you understand you have the tools. You understand how to how to effectively use prompts. You could get really good at at becoming a, as you call it, a prompt engineer, if you will. You're really good at that. So you have the tools, you have the prompts. You start using it. You have different applications. What are some of the challenges that you have faced as you continue to use this?
See, a few things you'll run into into into it would be so it's gonna be a lot of it's gonna be time consuming upfront in the beginning. As you are learning how to use a system, you might not have the most updated version of the system as well too. So it can be a little bit discouraging. It takes time for your IT team or digital team to keep up to keep up to pace because they're going to have to train the company data.
And so you have to be a little bit more patient in terms of, you know, sometimes there's going to be, like, bugs within it and it's just not going to like it's going to give you error messages over time. That gets a lot better. You again. And then so it's a lot of trial and error. As you become better and better with it, it becomes more fun. Because for me now, I'm like, oh, what is it actually gonna say?
And then so it becomes, again, it becomes, more exciting for me as in terms of, like, when I start, like, making as I'm making my client my prompts more and more complex. There are differences there. So ChatTpT is not the it's just the most popular language large language models out there. You will learn how to use the other sys the other, LLMs as well, too. So things like Claude, c l a u d e, that's another l m LLM out there. And then so it's, it's created by, the Anthropic Group.
And so what so if the child GPT is, like, a nerdy undergrad student, and then so it just, like, knows everything, and it's just gonna, like it's pretty quick to give you an answer. But Claude is will be he is able to think things more more complex problem. So it's like this under it is it's like this grad student in their last year, and then so they're, like, really smooth and, like, you don't have to give it as like, they understand your context.
And then so you'll learn how do you go between the differences the different elements and use it. So it's if you you do get better at it over time. You know, sometimes I have to, sometimes, you know, when I get I do a prompt and I don't get the the response, it can take me some time to just I have to, like, sit at sit for a couple of days, think about it, and then go back and then rewrite the prompt again. So it's it is like, you know, it's definitely a journey that you that you have to do.
It's not perfect. Right? We we just assume that we're gonna give it this this task. You're gonna give it this prompt, and then it's gonna be like, okay. It's gonna spit out the perfect thing. I think one of the biggest challenges, like, that most people I know face is, they they don't fully realize that there's a lot of there's still work that has to happen on your side.
Like you said before, you have to put in, like, 20% upfront and, like, maybe 20% on the back, and then it'll fill in a lot of the rest. But it's not gonna do the whole thing. You can't just say, okay. This needs to get done. I need you to do the whole thing without correcting or revising or summarizing it, or maybe even redoing the the prompt where it's like I because I've been in situations quite often where put a prompt in, I get the response, and I'm like, please try again.
Please add more options where you're just you continue to see what else it might be able to produce. Mhmm. And I'll I'll give you an example, Tom. So this is one thing that we do, like, very common as an MSL when we're trying to build our territory. And then sometimes, like, the last resort is going through, like, cold emails. And so if you and then, I mean, this is just a very simple task for ChatTP to do.
So, you know, write a custom write a customized email to this health care provider, and then it writes the email and you're like, it's just sounds like too long, too many niceties inside of there. And you're like, no human ever reads like or talks like this. So that's where it's like, you know, you it's, you're gonna end up in a loophole.
So that's where you kinda have to, like, think outside of the box about, you know, how am I going to I need to put certain restrictions or constrain the system and tell it exactly these things to follow. So I'm gonna do, I'm gonna read out that prompt in terms of, you know, writing and customizing writing a customized email to health care providers. And then, so here it is.
So I'm telling it to HeyCha GPT, refine the following email for me, and I give it the the the information about the provider. I'm not giving them a name, I'm just saying like, okay, this is a primary practice, family medicine doctor, he's been in practice for 20 years, you know, and then if I can grab some stuff about his bio, then I'll put it in there. So no names, you don't have to give it any specific names. Mhmm. And then here are the rules to follow when writing the email.
Use the f pattern of reading. So assume the reader will skim through the first words on the left hand side. Use active voice in the email. Short, simple sentences. Front load the important information of the sentences and then be concise. And then it finally, like, spits and I'm actually not using chat g gbt for it. I'm using a cloud model for this. Mhmm. And then it finally gives me something like, you know, like, this is something a human will read.
So it's adding a lot of more constraints into the system. And then so it's a lot I had to play around it with it for a little bit until I was able to get something. I'm like, this is what I wanted. And then so that is that's, that's an example of a good prompt. And and so that I use, and then you can make further refinements into it.
Of course, I'm always adding at the end, please suggest how I can improve my prompt, and it'll actually tell me, you know, what it intended to do, and what are ways to improve on the prompt. Well and that's what I was talking about. Like, that's exactly, I think, what, or a lot of that. First of all, that's a really good example that I think people are probably gonna rewind and listen to because that's that's such a great use case.
But to your point, and this is this is what I've had to learn, because I really didn't think I was very good at doing this in the beginning. I'm like, this isn't working. This isn't good for me. But you have to give it instructions. You have to say, I need you said the word concise. I need this email to be concise, or I need the email to be no more than 3 sentences Yep. Or a paragraph, or I need 3 paragraphs. This I'm doing a presentation, and the presentation I only have 8 minutes.
And here's my topic. Here's what I'm planning to include, but I need like, you have to give it really, really specific instructions, and then it works better, I have found. Yep. That's right. And then sometimes, I mean, the problem that I that I wrote, that comes from, like, email mark like marketing email, like, 101. So these are techniques of just how a human would read an email. So, like, you never start never start your email like, dear doctor, this is this. I hope this email finds you well.
Because if you're looking and you're reading, like, the the the header, and that's all they're gonna see in there, and then they're not gonna be like, okay. What is this email about? Garbage. They're not gonna even open up your email. Yeah. So it's like, use those things very precisely.
You're like, people only read the first sentences of the word, And so you wanna have the most important information upfront in the sentence and not all the way towards the end because you're just gonna skim through it. So it's like sometimes it requires you to pull things from, like, other industries and then in terms of, like, knowing how to write something. So it's just a constant refinement. Yeah. So how far into this AI journey are we for MSLs?
Is this something that, like, just you're doing, or is this something all MSLs are doing, or does it vary? Like, where are we? It's gonna vary. We I think we're still in the exploratory phases of this. It's in as a lot of, not a lot of companies, you know, are able to make these these Gen AI tools available to its employees. It is because it's so expensive. So certainly, you know, using it, you know, in your personal life and just gaining these skills will be helpful.
The way that, and then so the way I learned it was I am a busy working mom. I have a 3 and a half year old son, and I have a 1 year old, baby boy. And it was, like, pretty difficult for me to, like, you know, find a time throughout my day. I'm a busy MSL as well. And then so it's like, how do I, you know, incorporate this learning into, into my, into my week?
And so, you know, on my long drives to my ATP meetings, after I listened to your podcast, that's like on my Spotify library, I have lined up next to it, you know, an episode on prompt engineering, you know, updates on Microsoft Copilot, you know, scaling AI within a company. How do you, you know, convince, you know, how do you teach AI to your team members? So I have all these things that I'm listening to on the go, and then it really helps me a lot.
In the beginning, I had to listen to in the beginning, I would watch some YouTube. There's tons of YouTube videos in prompt engineering. But be beware, you know, if somebody's asking you, like, you need to download this specific prompt library, like, you don't need any of that, because the prompts are really more specific, to your role. And there's so there's tons of content out there on on, like, prompt engineering that you can teach yourself. So that's a great way to start.
If you, like, get the get the paid version of of either Chat GPT or Microsoft Copilot and start to use it. If you are you know, you have your own business that you're you're you're working on or you have, you know, certain, if you're taking a click a class you can use, those are perfect ways for you to use these tool to help you, and then you start practicing. I think gone are going to be the days where like in your resume, you're going to say you're proficient in Microsoft Word or Excel.
It's going to be like, I have proficiency in prompt engineering. I know how to use Microsoft Copilot. I know how to use, like, ChattyPT, Claude, like DALL E. Like, it's gonna be more like that. So it's, it's it's all it is. You know, it's the same thing we've learned how to use. We talk on the telephone, you know, where people thought it was just gonna be we or, like, how do you connect with somebody when you're just talking to them? I had you don't seem to see them.
But we learned and we adapted to it. We where, you know, everybody has, like, most people have an iPhone. And it's the same thing. You're gonna have these these tools, these JANIAAC tools. You're just gonna be able be able to incorporate on your day to day life. Yeah. So funny question or fun fact. I I actually treat chat GPT like it's, like, a person. So I'm very like, I'm, like, you know, thank you. Yeah. Awesome job. I actually I actually interact pretty well, and it responds. Yeah. Alright.
Do you treat the the the machine nicely? Yeah. Sometimes I'll do it. You know, I'll be like, take a moment to think about it. It's like, you know, take your time. Take your time. You're doing good. And, you know, for some reason, I don't know that actually, like, you know, behind the scene that actually makes a difference. But for some reason, like, I guess, you know, you give it other bit more time to think, it's going to, you know, give might give you a better response to it.
So the way these, like, large language models have been built is, you know, they were neurocognitive scientists or psychologists who, you know, work behind the scenes to build these, the neural network behind large language models. And then so, you know, that's that's why we're able to interact with it as a human. Right? So the way you talk is the way you wanna talk to the system. Right. And then, like yeah. And you're able to use to use voice prompts, voice prompts and tell it.
And so that's, you know yeah. That yeah. It's can treat it as a human, I think. Yeah. And it likes it. And who doesn't like to get compliments? Kindness always matters. That's my philosophy. Right? Yep. Yeah. Well, speaking of kindness, sent Nandene, you have been amazing. I this was this went so fast. I mean, we can keep talking, like, forever. So I might have to do a part 2 after you know, will you continue to perfect this and get better and better?
But I think a lot of people, they're gonna feel more comfortable jumping into it after listening to this. Yeah. And my goal would be, you know, for this to be, like, an MSL, like, community where we can talk about they can help each other along this, like, this AI route. You know, and so, you know, certainly this is a good start by having a conversation. You it's certainly a way to upscale yourself as an MSL. These these are gonna be important skills.
This is what our leadership team talks about. This is what they they're they're looking at these things, you know, especially within, like, research and development. And everybody is looking to see, like, how to, you know, how to incorporate this into our daily workflow. So having that having an MSL community where you can exchange information is really important.
You know, certainly, like, I've taken it to a little more deeper level and then to taking, so just being able to upskill myself, just taking a graduate level course on AI. And that helps me significantly in terms of, like, just being able to under interact with the system. Yeah. So you can, yeah, take it as far along as you want and go more deeper into it. And it gets more and more exciting. It I I know you're all excited. It sounds like it.
You you get, like, your face changes when you talk about it. Like, if for those of you guys listening, like, her face, like, is lit up like a like a jack o lantern. It's awesome. It's great to see. Well, thank you, my friend. I appreciate you coming on. This was awesome. And thank you, guys. Thank everyone for all your support, for listening to the podcast, for sharing the show. Always appreciate you guys, and we will see you next time.
Thank you, Tom. Bye. Bye. Thank you so much for listening to the show. If you've enjoyed it, please subscribe so that you don't miss episodes in the future, and feel free to leave a rating or a review or a comment. Thanks again, and I look forward to seeing you again soon.