What is Vets Who Code: Teaching veterans and leveraging AI

Speaker A: The thing that's been changing my life for, uh, the positive, giving me focus and, you know, keeping me busy and out of trouble, has been learning how to code. Right? It's a, uh, it's it. For me, it became the new boxing, you know, something to keep my mind sharp and, you know, build things, you know, using my hands. And I was like, well, why don't we just teach more veterans how to code? I was able to, you know, create this community, you know, not only get to, uh, understand code, write code, but get jobs as software engineers. Like, I was not an expert at this time by any means. Uh, I feel more expert now, but that was because I've failed more than most. And it was super. It was super cool.

Speaker B: I love Jerome Hardaway. He's a fantastic guy. He is the, uh, founder and executive director of Vets who Code, which is an organization that teaches vets how to code. Jerome, as a vet himself, he taught himself to become a full stack developer. Now he teaches others. I've known him for a really long time. I'm super excited to have him on the podcast because a, the work he does is really important, and B, I really want to know how he's using AI and how he's thinking about AI in kind of the modern way of teaching people to code. And that means when to use AI and when not to. And so we have a really great conversation about that. So check it out.

Speaker C: This podcast is sponsored by Google. Any remarks made by the speakers are their own and are not endorsed by Google.

Speaker B: Jerome, welcome to the podcast.

Speaker A: Thank you. Thank you for having me. And also thank you to the Google, uh, Gemini team for sending me all this cool, uh, Gemini swag.

Speaker B: I was going to say you were wearing the run kit. You're wearing some of the really cool swag, which is fantastic. It looks great on you. Um, so as I mentioned before, you and I, I think we've talked about this. I don't even know how long we followed each other, but probably a decade at least. And we finally met in person this year, which was really awesome meeting you at Google.

Speaker A: I o. And I was like, you look familiar. And you were like, yeah. I was like, oh, okay.

Speaker B: What's great about that, though, is that even though, yeah, you're meeting somebody in person for the first time, you do know who they are. You're like, okay, we followed each other for years. I know who Jerome is. Jerome is awesome. So actually, kind of speaking of that, let's talk a little bit about your origin story a little bit. And about the background around vets who code. So you were, you were in the Air Force, and, um, when you were in the Air Force, did you study, uh, did you do any work in it? Did you have any interest in computers, or was this all an interest that came after you left?

Speaker A: Roger that. Uh, I had no interest in computers. I didn't know anything about, um, computers, anything. I try. The funniest thing about my life is that I am consistently, continuously, um, Ms. Uh, like, mislabeled because of my. Because when you say Air Force, people automatically expect nerd. And when people look at my current background, the last 10, 11 years of my life, they expect me to say I was intel or I was an officer or, you know, I worked with computers in the Air Force, things of that nature. And I was like, no, I was security forces. I was Security Forces 3P 0X1. I was there with the, um, you know, making sure that the plane didn't, you know, get attacked, uh, or, uh, working with the military, working dogs and things of that nature. And people are like, wait, what? It's just a funny thing. Like I tell people, no, I was actually going to go to college for piano.

Speaker C: Ah.

Speaker A: I had a scholarship for Pacific University, and I was going to go do pursue music because I'm from Memphis originally. So, uh, it's very, uh, it's very hilarious.

Speaker B: Rock and roll.

Speaker C: Yeah.

Speaker B: Yeah. So, okay, so you're in security forces, you're in the Air Force, you leave the Air Force. So what did get you into tech? Especially if you didn't have an interest in computers beforehand? Like, what kind of sparked your interest and what made you decide, I guess, to settle down and go, okay, I want to do more with this and I want to become a software engineer.

Speaker A: So fast, uh, forward from 2004 to around 2010. I am, um, leaving the military, and it's the height of the Great Recession. Because of the things that were going on, I was not getting, uh, any opportunities. My military experience, I was getting, uh, thank you for your service, but we don't have any, um, we don't have any jobs for you. I'll never forget being turned down by the Department of Homeland Security, but only to get hired a year later for skills I picked up in software. For so many software engineering and coding. Um, but what really, you know, got my head into, you know, software engineering and coding was I saw this commercial, uh, code.org and I saw Bill Gates and Mark Zuckerberg talking on it, and I was like, okay, what are these nerds talking about? And then I went to Barnes and Noble, picked up a book on SQL or SQL or whatever y' all want to call it. You know, don't beat me up in the comments. Uh, and I taught myself, uh, SQL to the point where I could do Postgres. Really, really, uh, well. And I ended up getting hired for that by the Department of, uh, Homeland Security as a data analyst, where they're a certified TWIC program. So, you know, like I said, it came full circle where, uh, the military. The Department of Homeland Security didn't hire me for any skills I got in the military before, a skill that I picked up from a book a year later. And, you know, I was telling people, the amount of people who could, who were yelling at me or who could yell at me between, uh, from the military, civilian life was like down to four. It was always air conditioner. You know, it's not like that in Iraq and Afghanistan. And, you know, the amount of people who, uh, want to kick my butt is down to zero. So I'm like, I'm never going to leave. Plus the money looks great, right? So I'm here.

Speaker B: I love that. And actually, I mean, it's so fortuitous too. Like, so you got into Postgres, like, like Postgres right now is one of the most popular database. Everybody uses it. It's really had a moment in the last five or six years or so. Um, but like 2010, 2011, it wasn't, um, as popular as it is now. So, um, that was a, I think really great that for whatever reason you decided to pick up, um, data analysis. And B, it turned out to be really fortuitous because Postgres has become really popular. Like I said the last last few

Speaker A: years, the reason I picked Postgres is because DARPA had funded its creation, the creation of it with Stanford. So that was like a cool piece of military history with tech. And I was like, that's the one I'll, I'll pick. Right? So that's literally how I was. You know, when you're 20, this is how you make decisions.

Speaker B: Uh, and DARPA, for our listeners who aren't from the United States, is, um, resource organization that is responsible for the Internet and a bunch of other things.

Speaker A: Yeah. So DARPA is like the Department of the Army's research, uh, arm, Right. So they essentially, they. I mean, if they funded everything from GPS to the Internet to, uh, no touch or touch screen. So, you know, most of the things. You know, that's like one of the funniest things about Artificial intelligence. Now this is the first uh, tool that wasn't in the military first and then consumers got it. This is like, you know, consumers have it and now the military is getting it.

Speaker B: Yeah, that is funny. That's really interesting. Although I think like the military, like not, not generative AI and not like transformers, uh, but they have made investments, uh, you know, various countries and, and previous AI things. But, but yeah, you're right. This is kind of a funny moment. So, so you get this book at Barnes and Noble, you start to, to learn. Did you take any classes in person? Was this all self, um, taught stuff? How. What was your process?

Speaker A: It was, you know, it was self taught. So here's, you know, my process. I call it crawl, walk, run. And this is how not only, you know, how I teach myself, but how I teach everyone else. It's like you know, a progressive learning pattern. So I've always been this person where I am going to practice the thing and then I'm going to practice a new thing that piggybacks on a thing I just practiced so I can reinforce that learning of the, of the old thing while also keeping moving forward with the new thing and building and building and building on that so that to the point I can use all those things together. Right? So, and you know I'm, I m. Was just talking to some people earlier. I'm very big on writing and note taking and the way my brain works. If I write it down, um, and I, especially if I take like good notes and suddenly I don't need the notes. But if I don't take notes then out or I don't take great notes, then I'm not gonna remember anything.

Speaker B: Ok, so you're, you were kind of doing you know, a crawl, uh, walk, run, kind of learning from you know, one concept, then building on top of that and then building on top of that until you could do more and more. You get your job with the Department of Homeland Security. When did you decide that you wanted to do the nonprofit Vets who Code?

Speaker A: Uh, the nonprofit vetsu code was founded in 2014. So I was at, I was at uh, work playing. I actually was playing hooky. I took the day off because I was like, you know what? Uh, it's hot outside. I want to be outside of my friends. Uh, and then a tragedy happened back home. Uh, and essentially long story short, a veteran lost their life. And I was called, uh, by another friend of mine who was in the nonprofit space. She was telling me, hey, yo, so the way this has gone down the va has, you know, said they wouldn't help this person's family with burial. And because the VA has said it, all the nonprofits that are getting funding from the VA are also turning this family away. And I need your help with this. And I was like, what am I supposed to do? And she was like, I don't know, but I know that you can figure out and do something. So I was like, okay, well let's figure this out. I was wanting to break into software engineering anyway and I've been learning to teach myself how to code. HTML, CSS, JavaScript, Bootscrap. I uh, already knew the postgres, uh, so going to MySQL was not that big of a jump. And I know that I learned php, so I was like, you know, I'll just throw up a website and help, you know, try to help his family raise some money. And in that time, uh, took us about 27 hours, but we raised like 11, $13,000 was able to give that to the family and have them uh, you know, be able to give a proper burial. And then I was like, well, I'm done, goodbye. And people were like, wait, no, you can't just, you can't do something like that and then just go back into the shadows like a good natured vampire. And I was like, but that's what I want to do. So I sat down, I thought about it and I was like, well, the thing that's been changing my life, uh, for the positive, giving me focus and you know, keeping me busy and out of trouble, has been learning how to cope, right? It's it. For me, it became the new boxing. It was something to keep my mind sharp and you know, build things, you know, using my hands. And I was like, well, why don't we just teach more veterans how to code? And I never forget I was talking to people about that, about you know, teaching veterans how to code and can work, uh, uh, technically literate. And people were like, can veterans even learn how to code? It took my face in rooms where I was at. I was like, I am a veteran and I code. So I'm assuming yes, veterans do know how to uh, code and should be able to do this pretty, pretty efficiently. Um, but that was in 2014. Then I started just teaching people, like teaching people remotely because I was like, that was the best way in my, in my mind to be able to reach veterans that uh, would want to do this in a manner that would be worth my time not doing one offs and things of that nature. I was able to you know, create this community, talk, work through things, help people, workshop. And to the point where, you know, I got to the point where I was doing, like, 70, 80, 90 veterans at a time. Help them, you know, not only get to, uh, understand code, write code, but get jobs as software engineers and help them with their resumes. Uh, talking to them about this is how you should. I think this. Like, I was not an expert at this time by any means. Uh, I feel more expert now, but that just means I've failed more than most. And it was super, um. It was super cool. Super, uh, like, fun. Until the White house called in 2015, and they were like, so we've been hearing about this arm, this army dude that is teaching, uh, veterans how to code online. And I was like, well, I'm Air Force, so you might be talking about someone else. Like, oh, our bad. Yes, we did, Jerome. Yes, we are. We are talking about you. We thought you were army. I was like, nah, I'm not army at all. But thank you for miss, uh, servicing me. It's cool. Um, so I was invited to the White House for the first ever White House demo day. So they invited 50 Oryx. And I was the only one that was military and, uh, nonprofit to this. And I was like, holy crap. Like, uh, you know, it really put things in view because, like, as an enlisted troop, you see officers all the time, and you always had to salute them. But I was now in the White House, and the officers were opening doors and stuff for me. I was like, yo, this is kind of crazy. Like, how this is flipped. Being able to meet a president without, you know, having a bad incident happen is, you know, always cool. So after that, came back home, and, uh, this is when, uh, things really started blowing up and changing because, you know, one is a rarity. Have your work notice period to have your. Yeah, my work has always been very weird. It's. It's always noticed in reverse. Like, someone will notice my work nationally, and then locally, they'll notice my work. So I've never. I've never been the hometown hero. I've always been like, oh, draw. You just dropped this thing. Like a debut album. Like, no. Like, that was by accident. That's not what I wanted to do. Uh, so my first time, I went viral. My second time, you know, the White House, and I was just like, well, I'm just going to go back and do what I want to do. And people are like, no, you have to, you know, codify this. You have to, you know, make it more tangible and I was like, sure, cool. So that's when we started moving to everything to it's open source, having multiple cohorts. I tried to do two cohorts a year. Going from there. I started really focusing on just doing JavaScript and uh, Ruby because I saw the other code schools were doing that. But I started seeing around 2016, 20, well, 2015, 2016. A lot of computer science grads that became engineering managers were kind of hating on rubyus. I guess there's no other way to say it was, you know, oh, you know, the Ruby on Rails kits were making find it harder to get jobs because Ruby is made with programmer happiness in mind and Rails is made with kind of like our productivity in mind. So you can get a pretty good MVP or Rails API up, uh, pretty easy. Like they didn't really like that. So they started, you know, uh, essentially keeping kids who had like full stack developer because that was a indicator, uh, that you went to a boot camp. Uh, they were not hiring those troops, those kids, and they weren't also hiring kids or a lot of kids for Ruby. So that's when I was like, well, we're just going to drop Ruby and we're going to go all in on JavaScript. There's this cool thing uh, called Node that is already coming out. So we were one of the first orcs to go full marn stack, react, uh, Mongo, Node, uh, like we started doing all of that but after that, uh, you know, everything just started really exploding. My troops were getting jobs. I was getting to the point where we were like getting a 93, 97%, um, 20. 93, 90%, 97% each year success rate. And um, and let me just stop

Speaker B: you for a second. When you say success rate, do you mean that this is like the success, uh, rate of people who've gone through, you know, one of your, your boot camps, one of your courses and have then found another or how are you ranking? I guess, how are you defining success?

Speaker A: The only merit is getting a job in software engineering. Right? So it was one of those things that we were like, well, because we started seeing other organizations like, well, if you get any job with these skills or any space where it's like, you know, a certain dollar amount, then that counts. And we were like, nah, we taught you how to be coders, we're gonna help you become a coder. Um, and that's just how, you know, I always felt like I wanted to make sure that I was helping people reach the goals that they wanted to reach in I'M only able to raise funds based upon the amount of uh, troops, the storytelling, the success stories, people like, that's how it works. Donors like that is super awesome and I would love to help you continue doing this. The funniest thing that I love is when they hear how much it costs me to do some of the same outcomes or same scale of outcomes. And it takes like a hack reactor or something like that. Like, you know, they're like, wait, you've been doing this for 11 years, raising no more than $50,000 a year. I'm like, yeah, between partners like you know, like Google, y' all been um, a one partner for my work. Google, GitHub as well, like those type of technology partners, they really help offset the cost. Because of that, we've been able to keep that quality high and build like die hard, um, fans, people that buy our merchandise, people that uh, you know, come to our demo day every um, Veterans Day and see what we're doing new and talk to people that impacting the things that we're working on now are, you know, scarier and crazier than ever. Like we've been some of the things we're going to debut for our Vetsu code demo day. It's so funny. These are things that I brought up three years ago when AI first, first came to the scene that I would, that we would be able to do that people did not think were possible.

Speaker B: So like give me some examples.

Speaker A: So essentially we're going up and we're using or leveraging Gemini and our data to be able to score based upon their performance of how they do certain things, what they might need help on or what they might not need help on, what they might be ready for more advanced. Think of it like uh, going to leetcode and you go, hey yo, you know you're really, you know a Leetcode hard on Python while a, you know, leetcode easy for JavaScript. Let's go ahead and do this or you may need to go a little bit lower, need some more remedial on this thing. So let's go, go ahead and focus on this while we, you know we're going to push you up here, keep you down here. We're going to do that in real time. We're not going to wait until like a one on one or things like that. We're going to like I'm going to get the notification from email that I'm going to talk, uh, probably ping you let you know, um, make sure that you are, you know, that you're okay with this because we have to keep that human in the loop and then, you know, we'll move forward with that, with that lesson plan from there. Because I've had that problem with that with troops because they are super advanced in one thing. And I'm like, I'm teaching this person something. They're in a cohort and they should be doing something else. But I can't, you know, come and do something separate with them because.

Speaker B: Right, because. Because you can't scale yourself. Right. So. Because there's only one Jerome. But with AI, it's sounding like you can do more customized lesson plans in real time where it's able to kind of ascertain based on what, what their projects are doing and what the performance is to say. Okay, well, you're doing really well here. Maybe you don't need to continue with this lesson, but you can go on to this lesson. Maybe you need a little more work in area. Is that what I'm getting?

Speaker A: Yes, absolutely. That's.

Speaker B: Okay. Cool.

Speaker A: That's just one of the things. The funniest thing about one of the, some of the things that we're building is that I brought them up three years ago. Uh, people were like, uh, I don't know. This time I took a slower route and I was like, I'm going to wait till For Profit starts doing this so I can have case studies for when I do it. So people see that as more. Oh, that does make sense now. Like, you know, so there's a thing that we're, that we're bringing in for our, for our troops that is more focused that, that's already been used in, ah, education being used in higher ed. Uh, right now I'm actually a part of a team and at a company where I am, you know, going to be doing that work with higher education and colleges and doctors with interactive learning with artificial intelligence. It's very funny to me because I'm like, oh, I said that this stuff could happen and it's already happening in the for profit side, at the big colleges side, at the private school side. So now I get to bring it directly to the veterans.

Speaker B: Very cool. But I do want to ask you, I guess, just like to go back just to just a step when it comes to, I guess, you know, like AI content. Right. And how are you integrating, I guess, teaching AI tooling to your, your cohorts? Right. Because, you know, you talked about kind of like how, you know, you started out with like, you know, kind of like the, like the lamp stack Or Ruby, really, I guess, actually. Right. You were kind of, you know, teaching kind of that stack of stuff. And then technology changes, you move into, to react, you move into other frameworks. Now, AI is obviously a huge part of everything that is, um, going on. Um, so how do you then, I guess, um, start to introduce those concepts to your students, but also make sure that they're still learning the fundamentals that they'll need in day jobs and not just be regurgitating something to a prompt and not necessarily understanding, uh, you know, what code's coming out.

Speaker A: So this is very fun. So first, I don't treat AI in a. When it comes to education in a silo, a lot of people do that. They want to do AI separately and they don't, you know, they want to do this separate from the thing. And I feel like AI is embedded in each layer. Right. So we started the command line and, you know, like the best coding tools and resources out there getting pushed. Like, I love using Gemini, especially when it's like, um, Gemini Cli. When I'm using it, I like to use it to, uh. I'm using Claude squat with it. And I'm showing people how to use Claude Squad and Gemini and have like, agents. And I'm having. Oh, well, I'm having this agent check this agent's work. So I'm training this agent to look at this agent's work. And I teach the tool with like, you know, on top of what we're doing. So, yes, you're going to learn AI. We're going to use command line and AI. Right. So it's more about building new workflows, building new mind maps, and, uh, integrating AI into the workflow versus keeping it separate.

Speaker B: Okay.

Speaker A: One of the rules we have is if you use AI, you have to use Ask. You can't use Edit or Agent when you're learning, because that's something that I do. I don't use Agent or Edit when I'm learning.

Speaker B: That's really interesting. Now, do you get any pushback from folks about that? Um, because I agree with you, I think in principle, like, it is actually better, I think, for me to learn to do something the hard way rather than to use tools to go along with it. But, man, it's so easy to just let the agent do the work sometimes, right? And then, you know, I guess, do you get any pushback from folks about this or do you, like, how do you enforce this sort of.

Speaker A: I say hard, um, hard training, easy fight. Easy training, hard fight. Right. And I, uh, was like, so you're one. I'm not training you just to build things. I am training you for the entire job hunt cycle. Right. Uh, I think you and I, we've spoken this before. The hardest part of software engineering roles is getting your first software engineering job, right?

Speaker B: Yeah.

Speaker A: It's weird because you have access to all these tools that make productivity tools essentially you could get done in days, which just get done in weeks. But for the purpose of the interview, you can't use any of them. You're not going to be tackling this cold challenge with Internet and with uh, your uh, your co pilot or your Gemini. You're going to be doing this by yourself, just you off the dome. And that's what's going to make or break this. That helps them like, you know, come forward, being able to, you know, understand that. All right, this is why he's doing it. And also people want to become experts. Right? I, I think that's the best part about teacher veterans. Teacher people that, you know are military, coming, uh, from the military is they, they want to be their best. They don't want to do it easily. They want to do, uh, something's going to help flatten their, uh, their level of entry into tech. They'll do that hard work, they'll pay that tax. Right. Um, all I had to do was tell them why and explain to them and there's very, uh, few pushback. I'm not telling them to do something that I don't do. Like this is how I literally do things myself.

Speaker B: That was going to be another question I was going to ask you, is that, you know, obviously like you, you've become adept at uh, updating your coursework, you know, as the, you know, times change and as, you know, kind of the job markets change and whatnot. But with AI, there's, there's new stuff all the time. Like if, if you and I were talking this time a year ago, MCP did not exist. Agentic stuff was starting to become a thing. But like MCP as a concept was still a couple months out. Um, you know, kind of that full bore thing wasn't there? There were a lot of, um, you know, there were the thinking models. I don't even know if the first thinking models were even out yet. Right. So we were at like a very different place than we are even a year later. So it moves so quickly. How do you, I guess, keep up with all that stuff? And how frequently do you update, I guess, the content and what you're teaching

Speaker A: the AI wars or the AI updates reminds me of the JavaScript wars as I used to call them.

Speaker B: Yeah.

Speaker A: Where you know, every week there was a new JavaScript framework that dropped.

Speaker B: Yes.

Speaker A: And you didn't.

Speaker B: And then sometimes a meta framework. Sometimes a meta framework. Right. Because we need, because we had so many, we needed frameworks for our frameworks.

Speaker A: For me it's just like I'm, like I said, I'm a big reader. I focus on the thing I use, um, I do a couple of newsletters. I think the biggest newsletter I use for this is every uh. But I also, you know, I follow. I mean it's hard to not get the things from between Bite Bytes, Go and everything out there about MCP stuff. But you have to keep the main thing, the main thing when you're teaching. So I always focus on how close is this to the revenue line. What is the percentage of jobs asking for this? What is percentage of hiring, uh, managers or employers asking for exposure to this stuff. And that's where you know, the curriculum kind of gets where it start. I work my way backwards from the demand of the market versus from what the new high thing.

Speaker B: So you're looking at the trends in the hiring market. What, what are, what's, what's the job market looking for? And then you use that to inform um, your ah, your, your concept.

Speaker A: Yeah.

Speaker B: Okay.

Speaker A: And we use, we use AI with that. Right. So we're running, we're getting all this data and we're looking at doing percentages. Um, so I have a whole data pipeline set up that ingest this data, uh, we extract, we transform, we and Gemini as the next layer, uh, for that to be able to help us get our trends. And you know there are certain things that we see that you know, Python is SQL to top like they right now, people who have at least two to three years of experience in Python and SQL, they're having their job, uh, hunt essentially halved. Beyond that it's not so much looking at people that are strong and uh, their language is looking at people who are strong with and keeping up to date with their cloud stack. Right. So everybody is kind of becoming a cloud engineer that's using this type of uh, tool or a cloud engineer that does this. You're either AWS engineer or Google Engineer. I guess GCP would be the best term for that. GCP engineer or you're an Azure engineer and you're using those tools at scale to be able to do um, data pipeline, ctl, put products um, to market or AI will rag things of that nature.

Speaker B: Yeah.

Speaker A: So we have about 16, 17 hours worth of content on data just because of the job and what to look for things that nature.

Speaker B: Right.

Speaker A: Having those type of conversations, uh, or these type of processes really, like, it opens, I mean, it opens your eyes and like, it keeps, like I said, it keeps. The main thing. The main thing, you have to focus on the signal and not the noise. I really am focusing on helping, you know, helping the students critically think and, you know, navigate this world and like, you know, be calm, relax. I've done this before. I would. I remember, I remember when the boot camp craze popped off and then people started fighting against it by, uh, fire. By not hiring Ruby's. I remember the JavaScript wars. I've been, you know, I am old, Sarge. I've been through all of that. And then when you say all this stuff, people are like, dang. It really was a wild time to learn how to code in the early 2010s. Like, yeah, it was wild and people don't understand how, you know, like, you and I, we came up in the same era.

Speaker B: It was actually. It's funny to think about and. But it's a wild time now. But. No, but those are good lessons. Those are really good lessons. All right, I, I want to go on now to, uh, I guess some, some quickfire, uh, questions, um, for you. So I'm just going to ask you some, some real, real quick things. What, uh, was the last thing that you automated for yourself?

Speaker A: Last thing I automated myself was connecting all my kids schedules, including the adults. So that way I can have them on just one thing and they can add it and I can, I don't have to control it. I just see it on my phone so I know where people need me at.

Speaker B: I love that. I love that. What was the last thing that you asked Gemini?

Speaker A: Last thing I asked Gemini was to help me, uh, with an outline from my talk. Because I was like, I have, I can only give 20 minutes and I have 16 hours. Yeah, I was able to put up like a ton of resources in it.

Speaker B: Cool. Cool. All right. What, what is, ah, something you can do now that you couldn't do six months ago, man.

Speaker A: Uh, see, I think that's. I don't know.

Speaker B: Okay, well, let's, let's say a year ago then if six months is too, is too, uh, short.

Speaker A: Yeah, that actually makes us fly better because I was like, I've been on the AI train for a minute, right? So three years feels like forever. Four, uh, years feel like forever. So thing I can do now is when we talk about, when we talk about mcp, uh, MCPS and being able to connect those. That's something that I can, that I can do now and leverage with all of my tools. But the thing with AI now is, um, I think because of my, because my expertise with programming and with AI, I'm spending less money on tools because. Or I'm using less tools because I don't need them. Because now I can build like a tool for one. Like, I can now have my. I have a process where I have my uh, you know, my AI, uh, my AI note taker, and then I can also have it coming directly to my Slack and then I'm adding and editing directly in Slack. I'm doing all these things, uh, from this one area now because of these, uh, connectors of the, of these applications and AI and things of that nature. So I feel like that, I mean, spending less money to do more is the thing that I'm doing now.

Speaker B: That's awesome. I was going to say now. Now how have your API bills gone up? Because mine have, Mine have not.

Speaker A: Like, that's what I'm saying. I, I mean the way I, the way I use AI is very different from people. Like, I'm a big model switcher and I like the model that uh, like Claude only reviews code. Um, Gemini is the one that's writing the code, but Claude is the code reviewer. And that keeps the, that keeps the cost down. Especially if I'm doing like, if I'm doing, following the software engineering pipeline rules of, you know, doing it, you know, one branch at a time, one ticket, one problem at a time. Like it just, it keeps everything at bay. And I, you know, I'm managing uh, Gemini as the uh, coder and making sure, like working with it. Then once we pushed it to uh, to GitHub and we push it to, you know, get a PR, I'm managing, uh, the co founder mentioned Claude and Copilot to review it. And then like, that is, that's just my process and it has not impacted, like, it hasn't impacted my spin at all. But you want to know what's the real sleeper in AI that people aren't talking about? JetBrains. JetBrains IDE for Pycharm when you use like Gemini Cli and I have my Gemini code, uh, attached to it. It is a real, like Gemini and pycharm have been working like wonders for me and I think people are sleeping on using Gemini with pycharm. When you combine Gemini with the intellisense that, uh, Jetbrains already has. You're getting a wonderful experience and people are really, People are really sleeping on that. I don't hear. I don't hear enough about the Gemini and Jet Brain stuff. I hear barely anything about it. I'm like one of three people that. I'm like, oh, no, when you put that Google Gemini layer right on top of that, uh, that JetBrains, uh, IntelliSense, or like that, it's a wonderful experience.

Speaker B: Fantastic. I love that. I love that. All right, so final question is going to be, um, I guess, uh, and you talked about this a little bit, but, like, what. What's the next set of problems you're trying to solve?

Speaker A: Next set of problems I'm trying to solve is how to help bring to the masses about the changes in the workforce and how to solve them. Right. So I have 16 hours of data. I've been doing a lot of research. Been. I've. I know every layer and section of getting, uh, a job as software engineer in this era with AI. But it's just like, how do I format them? How do I get this to write people, which verticals? Because when I say data and I like content of this only. Oh, this is for a different vertical. Like, people who write JavaScript require different things and people that do DevOps. People who DevOps require different things and people who do, uh, ML or data engineering and things like that. Right. So, you know, it's. It's so many moving things. I'm like, how do I, how do I help everyone? Is the thing that I'm, like, looking for. And like, that's so that's like, the problem I'm trying to solve.

Speaker B: I love that. So basically what we need to do is we need to create like a, A Jerome agent, right. So that, so that everybody can have like a, uh, you know, a Jerome, like in their cli or in their favorite tool or whatever the case may be, and help them along.

Speaker A: A Python drone.

Speaker B: Yes, right, exactly. That's what I'm saying. Like, have one for each of these things. I would love that. Yeah. All right, Jerome, thank you so much for taking the time to talk with me. We will have links in the show notes where people can go to learn more about, uh, Vets who Code. But, uh, if you want to give us a plug for the website or the GitHub or where people should go if they want to learn more, if they want to donate.

Speaker A: Roger that. So three places we love when it comes to Vets who code is our YouTube channel. Uh, Vetsu code, uh, YouTube, uh, that's code LinkedIn and of course VetsuCode IO and you can, uh, donate from anywhere on the site with a donation page.

Speaker B: Awesome. Awesome. Well, thank you. Thank you for all the work that you're doing. Thank you for helping so many people learn and get skilled up and get jobs. I think this is what you do is incredible. I think you're an incredible person and uh, I can't wait to see the Jerome, um, um, Agent hit the streets.

Speaker A: Oh man, that'd be the day. My wife is like, all we need is more versions of you. I can hear her voice right now.

Speaker B: Thank you so much, Jerome.

Speaker A: No, thank you. Thank you.

Speaker C: Thanks for joining us. For more information about our guests today, please check out the description.

Speaker B: And for more great conversations like this, hit that subscribe button. Until next time, thanks for listening.