Stop Scrolling, Start Building: Vibe Coding, AI Safety, and Tech That Calms You Down

I know we all have a love-hate relationship with technology. I'm a technophile, but at the same time, I understand that it has to be used in a very mindful way. And that is why I'm so excited to have our guest here today, Gabe Montague. He is the founder of Mindful Makers, a Boston-based community exploring how technology and AI can support mindfulness rather than erode it. Hello and welcome to the Mind Body Space Podcast. I'm your host, Dr. Juna Bobby MD. I'm a double board certified radiologist and lifestyle medicine physician and a Harvard trained educator. And this is the Mind Body Space Podcast where we talk about the science of resilience. I've been working really hard on publishing Relax to Soar. This is not just another planner, it is actually a resilient system, and it's on paper for a reason. Our brains respond so much better when we actually write things down on paper, and this system calms stress, beats procrastination, and builds the microhabits that quietly change everything. Because we all know what to do, but most of my students over the years and myself have experienced that feeling of overwhelm. Relax to Soar system allows you to change at the cellular levels so that you can experience optimal recovery and more follow-through. That means action towards the goals that you want to achieve in life. INSIDE is a shortcut to all of the classes that I've taught over the last 10 years at places like Juilliard School, Columbia Law School, NYU Medical School, and to thousands of people online. I needed this myself. This has all the components of lifestyle medicine and habit setting to increase your mental and physical resilience down to the cellular level. We're having a special right now, so head on over to mindbodyspace.com/inside. Store. For a limited time, you're gonna get free coaching from me to help you get a head start on your goals. I'm so excited to have here today Gabe Montague. He's the founder of Mindful Makers, a Boston-based community exploring how technology and AI can support mindfulness rather than erode it. The group runs biweekly, right? Every other Saturday. And it's a hackathon. It has intentional tech creation, and Gabe's gonna tell us a lot more about it. The reason I love Gabe is because he is so concerned, and rightly so, about AI and the interaction with humans as well as security issues and ethical use of AI. So Gabe, you went to Harvard, you studied computer science there, and you were on the track and field team. So maybe you could tell us first a little bit about Mindful Makers, why you started it. And that's how I met you because I, I went to your first hackathon, which was so cool. So maybe you could tell us about that and then we'll go backwards on what you did after school. It seems like you had a lot of projects that you worked on, right?

Well, thanks for having me.

So tell us, first of all, like, what is a hackathon, right? For people who are listening, maybe they don't even know what that is.

Exactly. Yeah.

Sounds cool.

What is a hackathon? So a hackathon used to be something really scary and not that good for you, which is, or maybe not scary, but intimidating, which is like, 3 days, at least my first hackathon was at MIT. We spent 3 days, a bunch of people gathered and a company sponsored a bunch of ideas and kind of divvied them out to teams and the teams picked them up and they would compete for 3 days, get very little sleep, maybe drive each other crazy, maybe get along and get a lot of work done. And then at the end everyone would present and usually there's some prize money. So that's, that's I think a stereotypical idea of a hackathon.

So that was your first hackathon at MIT. What did you work on when you were there? And did you have your partners with you, or did they assign you, or did you break out into partners?

I think the serious teams brought their partners with them, but I just actually was just walking by on the street and someone like recruited me on the street. It was a crazy way to join a hackathon.

That's often how things go up there, right? You just run into so many interesting things going on all over Cambridge and you just end up in like a conference or a hackathon. So cool. Yeah.

Yeah. So I showed up to the info session at the start of the hackathon without my laptop and decided, you know what, I've heard about these things for years. I might as well try one. And yeah, we were working on some AI research agents for researching which kinds of bioreactors were best for producing which kinds of crops. So it was a very, like, commercial—

Very MIT. Yeah.

Yeah. It was very commercial because it clearly was going to help this company a lot.

So very cool. So back in those days, It was like 3 days in a row. You weren't even sleeping. You guys just kept working through the night. They gave you donuts and pizza. Is that what happened?

Well, I think some teams didn't sleep, but we weren't kidding ourselves about, like, I don't know. I don't think we thought we were going to win, so we slept.

So hackathon, like a bunch of people come together, you brainstorm ideas. Maybe there's like, in your case, Mindful Makers, which you'll tell us about.

Right.

You have like a guiding format, but so then people come together, sometimes they pitch each other and try to get each other on a team, or you come with a team and then you create an actual product or like a beta, usually a beta version of the product, right? Or something that's just like a minimal viable product.

Right. Yeah. So I guess so far I haven't given a very good optics on, or I haven't sold hackathons very well. So, but, but these days they're very different and Mindful Makers does a different kind of hackathon than the first one I went to. Because after that, I thought that was really fun, but I'm never going to do that again. And so, but what we do is we get together just for one day. It wasn't until later that I realized you can do a hackathon, especially now that AI is out, that vibe coding has, has become a thing. You can do it just in one day and you even need, sometimes it can be as short as like 5 hours or 4 hours, just a handful of hours is, is sometimes enough to create something awesome and to prototype it. Because AI has accelerated the development of coding so much. So that's— so Mindful Makers is a group that comes together. We get coders, we get non-coders, and we try to take care of each other's digital spaces in a way that's conducive to mindfulness and meditation. So just like you would organize your room at your house in a way that is designed to make you happy, our digital lives, our digital spaces, we don't have a lot of control over.. And arranging these digital spaces with technology that's in the best interest of our mental health and our wellbeing is, is what we try to do in the group and takes a lot of different forms. But ultimately we're, we're promoting mindfulness in the daily lives of, of our members and beyond with, with Mindful Makers.

And like using tech to become mindful or being mindful of your tech use, right?

Right. It's kind of both.

Yeah. I totally need help with. I was just complaining to my therapist. I'm using my phone way too much. So that was like a problem I had. So I need your help.

Everyone does.

But hackathons, believe it or not. So I was giving a few talks recently, and I know you can't believe this, but some people have not heard of what Vibe Coding is. So maybe you can explain that. I know the first year that I was, you know, when I was back, I went back to school recently. So I was at MIT and they said, this is the first year we're offering this AI course to people who are non-coders because finally AI is able to help you. Just, you could talk to AI and build things. So one of the classes that I took, they, it was the first year that they offered it to people who don't know how to code at all, which is how I got in there. So tell us, what is Vibe Coding and how do you do it? If somebody was gonna do it today, 'cause you can, right? Somebody could just start doing it today.

Yeah. Vibe Coding is using AI to write code. So people are familiar maybe with going to chatgpt.com and typing in a question to ChatGPT. It's very similar to that. You have a place where you can communicate with an agent, except instead of asking the agents to write a paper for you or something like that, you would ask it to write you a website. And then you see the website right there next to your chat. And that's pretty much how it works. I think with Claude or— I don't use Claude or OpenAI, but I think they can do that, right? Especially Claude, I think can do that right in Claude these days.

Wait, I thought you used Claude for coding.

I use Claude Code. Sorry, I meant vanilla Claude. I think even the, the consumer-facing entry point for Anthropic or for Claude can, can do this already.

So you use the coding version, which is Claude Code, and I, I have Claude regular and I have Claude Cowork. And Claude Cowork, you can actually tell it to build you an entire website or just by saying use these colors. You could even say something like use Harvard colors and MIT colors and create a website for me that looks technical or looks more aesthetically pleasing or— and just talk to it like a person, right?

Yeah. Yeah.

Maybe even more like a person than a coding person. Yeah.

Yeah. Who understands the fact that you don't understand, you know, if you're, if you're not technical, you don't know what— I don't know. Breadcrumbs are on a website, that doesn't matter. You say the thing that lets you go back to the page you were on before and it's kind of horizontally aligned, it will understand you, you know?

Yes. Yes. So you talk to more like a person than you would to a person who's a coder, maybe.

Yeah, I think you're right about that.

So if we go like step by step, so you would download Claude and then you would say, what would you say to it? What's an example of something you could build? Yeah.

So, so when I teach wide coding, I, I oftentimes point them to Lovable. So that's what I know. I don't actually know if the capabilities of Claude right out of the website. Uh, Lovable, you just go to Lovable's website and then you think what you want to make. It could even be like a PowerPoint, I think, or a brochure or something. It could be static and for the purposes of printing out, or it could be a website that you're gonna hand to your friends or pitch as a product that is in development, but you just tell it what you want. And it goes and it builds it and it's probably going to have broken privacy policy link and stuff like that. And images that look generated and a font with a gradient on it that is a telltale sign of AI. Yeah. But it looks better than a non-coder could ever hope to, to make.

And it actually works, seriously, right? So I built something on Lovable. One of the first things I vibe coded was a while back, over a year ago, and I just asked it to, used, you know, my stuffed animal that I have, the plush doll, the Brain Bun, and I gave it a photo of Brain Bun and its friend Zen Cloud, and I said, create a neuroscience website that kids could talk to. I mean, I gave it more details than that and I kept talking back and forth about things I wanted to change, but it came up with really cute things. It had Brain Bun floating kind of on the screen. And then nice. I said, make it into a chat thing so you can chat with it as if you were texting. So you'd say, oh, I'm upset. And then BrainBun would say, well, your emotional brain is doing this and it would explain it. So that's what I coded on Lovable. And most recently with Claude, I was able to code a sleep app where it walked through all the evidence back steps to get you to go to sleep. So you would just check off boxes. It's still on my website.

I think I saw that. Yeah. Yeah.

So that was really cool because that, I thought was so much easier to do than what I did on Lovable over a year and a half ago. And you just keep talking to it. The only problem is if you're like me and you're not very clear in your communication, it also gets very confused at times. So you have to start up all over again from scratch. Now I started talking to, let's say, ChatGPT or Claude, then getting Chat or Claude to give me the instructions. To give to Claude to make the thing.

That way it knows what breadcrumbs are, to use that earlier example, and it can put that in there and it doesn't have to waste time figuring out what you meant. What, you know?

Yes. Breadcrumbs.

Yeah, that's a very good strategy.

Breadcrumbs, which is what placeholders, so the program knows where you were, right? Like what you're clicking on.

Yeah, exactly. Or radio buttons. Yeah. Any, anything, a hamburger menu.

Yeah. So now you have people who can't code. But when people show up at your hackathon, most people do kind of know computer science, right?

No, I just ask that they are not actively afraid of technology, that they bring their laptop and their willingness to learn.

Actively afraid, the phrase, I love that.

Some people say, ooh no.

What about passively afraid?

Yeah, I mean, they can be afraid but willing to learn, you know, and some people have had great success. Like at our last hackathon, I was very proud of someone who came last time and was a little bit veering towards that actively afraid of technology type, but she made an app in Lovable and figured it out. You know, it's, it's really not that hard once you overcome the fear of typing in something. I guess the hardest part is realizing what's possible in a website versus what's not possible in a website.

What was the coolest thing somebody's made so far in Mindful Makers?

Oh, the coolest thing that someone's made so far. Mm. There have been lots of cool things. Some people were working on telepathy.

Is that proven to be true? I mean, well, signals.

Yeah, very loose version where two people put on EEG devices on their heads and it transfers, uh, audio waves like binaural beats in between them so that it actually is synchronizing their brainwave frequencies actively.

That's so cool. Wait, that's very cool.

That's on the more experimental end of things that we work on, but the idea is that it promotes connection. We kind of ask like, how is this going to help mindfulness or being in the moment for someone here in the room? And that's kind of the criteria of, of what we try to make.

You're trying to connect people through technology.

Yeah, connecting people is, is one way, or even just having them land more in the present as they go about their daily life.

Oh, that's so interesting because we had this toy 10 years ago when my kids were little. It's like you put this thing on your head and it picks up the electrical signals from your brain. You can float little balls.

Wow.

I'm going to have to send it to you.

Yeah.

But that brings me to say that your Mindful Makers Hackathon, a lot of times you have some companies join you, right, that are in the neurotech space. Yeah. So can you tell us a little bit about that and how that works in the hackathon?

Yeah. So EEG is a popular emerging technology. It can't read every thought you have, but it's getting better and better. There are models developed at MIT that can correlate your EEG signals with different things that you might be thinking of at the time. Those generally require really expensive hardware that we don't bring to Hacks. So we're operating more at the level often of an EEG being able to measure how stressed we are or how excited or how sad or I mean, various emotions correlate with different EEG signals, some well, some not. But each, each brain state does have an electrical signature that these machines that you strap onto your head can, can measure. And some of them, you know, are things that you would never walk around with in daily life and some of them actually are very unobtrusive. So it's a new kind of wearable cap that goes, right? Yeah. Yeah. There's a cap for our new next Next weekend on the 30th, we have a hackathon with an EEG company that just goes behind your ear. So you don't even realize that you have this wearable that's monitoring your brain. It's very cutting edge. It's not something that I would use during my meditation necessarily, but it's something that a lot of people do use as part of their meditation actually. Like the Muse headband is probably the most famous example of this.

Yes, I used to use the Muse headband at Juilliard with all the musicians there. And when I was teaching them how to Elicit the relaxation response, which is the opposite of the stress response. But EEGs, I would say my neurology doctor friends would say that a lot of these wearables are not picking up all of the signals. It's just basically like the muse is in the frontal lobe, which is where you want to increase signal when you're meditating. And so we call it oscillations and your brain does have a signature oscillation. And that is basically the electrical signals you're picking up from different parts of your brain. And you can tell when someone's like sleepy, things like that. But it can't really— I mean, the thought reading thing, I did read about it. I can't remember exactly how the mechanism is. It's not like it's reading your thoughts. It's actually like translating whatever electrical signals into possible words. I can't remember exactly how it works, but yeah, I don't get up again.

I don't know.

But it's not like they're reading your thoughts, right?

Right.

But they can tell if you're getting drowsy or not. So that's interesting. And there was an article that talks about how employers might know if you're getting drowsy or if you're getting distracted through earphones that you could wear.

Yeah, that's, that's what we try not to build at Mindful Makers. I mean, but ultimately technology that could be used for good can be used for, I know, corporate spyware at the same time. We were just talking about this at the last—

corporate spyware.

Yeah. At the last hack, someone made a device where when you get up, because the idea is practicing gratitude is shown to help our wellbeing, our happiness, our sense of connection. But oftentimes these, these apps that just interrupt your daily flow, become annoying. You're right in the middle of something and then your computer says, oh, think of 3 things you're grateful for. The team at The Last Hack built hardware that is with a proximity sensor that measures and knows when you get up from your desk and that's when it pings your phone to remember these things. Um, oh, okay. So yeah, we were joking.

That's, that's cool.

Yeah.

So it has a sensor that picks up the fact that you're getting up out of your desk. So instead of interrupting you during your focused work session, it knows when you're about to get up to go to the bathroom or get a drink of water, and then it says, think of something grateful that you're grateful for.

That's cool.

I like that. Yeah, I like that.

I think hardware is somewhere where I'm most excited to explore more with Mindful Makers. Our screens are such high-value real estate for so many companies that don't have our best interests at heart. But when we introduce these new devices, I think a lot of people get a lot of value.

So you studied computer science and AI, but you also— are you a mechanical engineer? Because I saw that in your bio you did some engineering or like hardware things.

Yeah, yeah.

Can you tell us a little bit?

I didn't say I did it well, but—

You're too modest. Well, you're experimenting, right? So tell us what you made or what you did.

Yeah, sure. So we were working on micromobility technology, and micromobility is basically bikes and scooters and then 2019 and 2018, there was a wave of interest in electric bikes and electric scooters.

Mm-hmm.

So Metromobility, the company I started at that time, was making custom hardware and I was trying to learn how to use a solder iron. I never studied that in undergrad. Yeah. But suddenly we were trying to make these electric bikes and charging infrastructure, weatherproofing it. But luckily I had lots of talented people. That know a lot more about that than I do, taking care of the hardware at Microlearning.

So it sounds like you've always been very creative with technology and trying to make new things that don't exist.

Yeah. Well, yeah, I mean, even before Mindful Makers, I was developing kind of this technology on my own, especially during the pandemic when there wasn't a lot of us to do.

Which technology?

So for instance, I think during the pandemic I started work on an operating system where you would have to tell it what you're going to work on before it unlocks anything else about your computer. So you have to start with an intention. The— I think the most popular version of this that I see is people that disable suggested content on YouTube. So then they go to YouTube, but then they have to intentionally type in something that they want to see into the search bar. Okay. I took this further and tried to make my whole computer like that, um, before AI. So, uh, and then of course, uh, there's an aspect of, of of seeing your data, seeing how you spend your time as being a helpful indicator for how you want to spend tomorrow. So it had this tracking element as well.

So if you like opened your computer, you'd have to say, I am going to work on my book today. For example, I'm writing a book and I get so distracted. So I would type in, I'm gonna write my book today.

Right.

And then it would not allow me to open anything except for what I needed for that task. Is that what you're saying?

Yep. And then that was before AI. So I need that.

So now you can do that with Claude, right?

Now it's a lot easier.

But I'm scared to put Claude on my desktop. I still haven't done it because I downloaded it and everything. And then it said, check off this box saying that you read this and treat Claude agent like it's a person that you're employing and make sure you get to know it and don't allow it to have access to everything. It said all this stuff like, Yeah, we don't know what it might do, things that are unpredictable and like, just be careful. So I got kind of spooked and I didn't do it, but I, I'm still tempted to try it out soon. What do you think about, what do you think about this new Claude desktop thing?

I use it a lot.

Yeah, you do?

I, I use it a lot.

You gave it certain privileges to certain folders, but not your whole desktop, right?

No, I pretty much give it access to everything.

All right. All right. If you're doing it, I'm gonna do it too.

I think it's a matter of I do that when I'm watching it. I think they, they put that in the terms of service because they want you to think of it as a new intern that you just hired at your company.

Yes. That's what it says.

Something like that. Yeah. Yeah. And you wouldn't just give your intern access to your bank account.

Yeah.

Or your PayPal. And tell it to, you know, generate billing statements and then walk out the door and leave it going.

So you watch it, you don't leave it. So you can see it like going around on your computer, clicking on things and opening folders and things like that.

Actually, I don't let it click on things cuz that really could access my, you know, like a bank account for instance. I, I usually let it go around and access folders on my computer.

And did you allow it to have access to all folders or just like toggle certain folders? You can tell it to only have certain folders, right?

Right. So I, I do all folders, but I know Unix permissions and security, so I lock down, you know, what it can do, but not using the mechanism that Claude provides.

Oh, so you go into your Unix or your terminal?

Yeah. Yeah.

I get so scared every time terminal opens on my computer.

I'm like, whoa. Yeah.

Yeah. I know I shouldn't be scared, but you know, when the terminal opens for us kind of non-CS people.

Right. Right. That's a very intimidating— yeah. And then—

yes, the black terminal. Although when I was in a lab, I did neuroscience research before and during medical school. And I would always work in Terminal.

So, oh, nice.

I do remember it a little bit. Yeah. But we just did very little with it. But that's so exciting. I just feel like I'm not organized enough to use Claude in the most effective way possible. You know what I mean?

I think everyone feels that way.

Oh, really?

I mean, I think the only people who don't feel that way are the ones who are giving it way too much permission and are not being careful enough. At least it was like that back when I think OpenClaw first came out.

OpenClaw, which is an open software operating system or software or AI.

Yeah, this was an open source tool that just prompted the user to go and download other AI tools without much thought about safety or guardrails. And it spooked me a little bit. How easy it was.

It did amazing things. And so everybody was downloading it, but then there was some security issues, right?

Yep.

As usual.

There was some.

So I guess what would you recommend then? Just keep maybe all your sensitive financial stuff off of your major laptop that you're using a lot of AI on. Would you recommend that?

So when I last was looking at this was when OpenClaw was just coming out. And a ridiculous amount of its library that it drew from for capabilities was straight up malware. I think it was over, I don't remember the percentage, but definitely over 1%, maybe over 5% was just malware that would sell your data, including maybe passwords. Oh, okay. So it was a landmine. Um, and these days I think OpenAI and Claude are trying to get into this area of they, they want it to be useful, but they also want to protect their users. So I think a good strategy is just sticking with the curve of what OpenAI and Claude allow in their desktop apps, unless you're ready to do some extra locking down and do some extra thinking about security. Like you could buy another device.

OpenAI is like free and it's like open. It's, it's not like paid like Claude and OpenAI. So that's what you're talking about for people who don't know OpenClaw, right?

Yeah. Yeah. It's like OpenClaw is kind of oriented towards somewhat developers. At least it was for a while there. Mm-hmm. And developers have a little bit more sense of what's going on when they unleash an AI like that on their system. But it still was very dangerous for a long time. Yeah.

Mm-hmm. Okay. So for people who are listening who are not like CS people or they have kids or whatever. Hackathons, I feel, is something fun that you can maybe even bring to high schoolers, or, you know, certainly college kids do it, but maybe even younger kids. So you and I discussed this, like, it's something that could be fun for younger kids, and, and especially because AI is so prevalent now, people should really learn how to use it. You think— do you think young kids should be using AI right now?

Oh, that's a good question. I mean, there are a whole bunch of people who do research on this, and I don't—

What do you think though?

What do I think?

Yeah.

I think with ChatGPT, there's a danger of the sycophancy that can drive people towards, I don't know, destructive behavior. So that's something that I'd be worried about. But I think when it comes to vibe coding, it's different. Vibe coding is not about forming an attachment to the coding agent. It's about asking and dreaming and thinking about what you want and, and designing something. And it's, it's very powerful. It's how I started Mindful Makers. I started Mindful Makers in Lovable. I just had the idea and I typed it in and then suddenly it was a website and then suddenly I could show it to people. And that's really what the point is of rapid prototyping is you, you kind of can sell your idea better with vibe coding, even if it doesn't exist yet. So it is very powerful.

Lovable. You can literally make an app, you can deploy it, and you can have up to, I think, 1,000 users or something. I think there's a limit. Last I checked, there was, but you can actually deploy it on. Yeah.

And you can make your own ChatGPT in Lovable and it would work.

Yeah. I think that kids should definitely use AI.

Yeah. Yeah.

They should learn to use AI like any other tool with, um, somebody watching over them and making sure that they're using it in the right way.

Right.

But it's just like when computers came out or the internet, you don't wanna let them loose with it in their bedrooms with the door closed. Yeah. But it's something that everybody, I think, needs to learn.

Yeah. I totally agree with that.

It's not going anywhere right now. Some people say it could dumb you down. Others say it could speed you up. I think both are true. What do you think?

It definitely is dumbing me down in my ability to write code, right? I'm writing less and less code and asking more and more about things that I want to see in the code. I'm also learning a lot. I'm learning new, new coding technologies a lot faster than before, though. So I've, you know, in terms of the actual encyclopedic knowledge, it helps a lot for like I can just ask about any programming concept. What is this? I was asking about encryption algorithms this morning.

Encryption, which has to do with security.

Yeah, right.

But when you say losing coding skills, You already know how to code, so you can actually go in there and edit it. You're more like an editor now rather than the writer of code.

Right.

But it's not like you're writing ideas, it's more code. So does that take away from the logic of coding that you used to have, you think? Or you don't need to have that logical— I don't know. I think it helps to know code, right? To use it.

It does help, but I'm not too worried about it. It's like I learned how to do integrals in calculus class in high school. And then I haven't, as a computer science major, had to do that many integrals. And then I think that's fine. I could learn again if I needed to.

But don't you think the integrals helped you, like learning that helped you think a certain way though?

Yeah, I'm glad I did.

So it's not like it's useless even though you might not be using it daily.

Yeah. So what I would recommend is asking ChatGPT first, how could I write an app and not using Lovable right away. Lovable does too much at first. So I see people really learn the best when they're saying, I want to type everything in every line of code in myself. And you tell me what it means and why I should put it there. And that's a really fast way to learn, especially when you, yeah, you can pair that. I was actually teaching vibe coding to middle schoolers in Framingham.

Oh, when did you start that?

Framingham. That was because one of my friends from the Harvard Ed School had a program doing this and I thought it was fun and I volunteered, you know, awesome to, to help her out with it. And it was a lot of fun.

So you had them talk to AI to teach them how to write the code line by line and then tell them what that meant, that line.

I mean, it was kind of an entrepreneurship-focused class, which I have never heard of for middle schoolers. So it was very unique.

Oh, I love that.

But yeah, they weren't all interested in learning how to code, but what they did all have was something that made them passionate. And they all had the ability to ask AI, how do I go about making this a thing? And then they started teaching them, whether it was the one kid who wanted to do a robotics project or other kids that wanted to make a game or other kids that wanted to make an app. ChatGPT started teaching them the technology and they all started learning, like they all were attached to their idea. It was their idea. It wasn't forced upon them like integrals in calculus class.

Mm-hmm.

And it really worked.

I, I think— And then what did they come up with at the end of the class? Did they have like a little MVP, minimal viable product?

Yeah. Yeah. They each had their own little MVP. I mean, that's one group made a game, one group made a robot, and one group made an app, I think. Yeah.

And they used the chat for that.

Yeah. I think one of the groups used Lovable more, and then one of them was just using Cursor, which is it was harder to set up, but they figured it out.

Cursor is another type of like coding app.

Yeah. Except you download it, but it's not that different.

I think kids should be exposed to it in a controlled situation.

Yeah.

With supervision. And in some ways it's, you know, sometimes when it's so prolifically spitting things out at me, then I feel like I am getting dumb. 'Cause I, I feel like, oh my gosh, I can't even like review all this stuff. So what I've started to do is like say, ask me before you go to the next step.

Mm-hmm. Yeah.

So that I could digest it, I could edit it, I could have more control over what it's gonna give me in the end. Because sometimes it'll just give you so much output that you just say, you know what, I'm not even gonna look at that.

Right, right.

It's too much. And then you have a computer littered with stuff. And I'm trying to learn how to do hygiene with my computer now. So as you can see, my environment is very— I don't have that much stuff. I used to have a lot more stuff. Minimalism is something I love, but my computer is horrendous right now.

But at least I know.

Well, that made me feel better. But there's that comedian Jim Gaffigan. Do you know him?

No.

He's hysterical. But anyway, he talks about how he doesn't know how to clean up his computer, so he buys a computer every couple of years because it's like, oh, that was 2024, that's 2025. I'm getting to that at this point. So any tips on hygiene? Just getting rid of stuff as soon as you don't use it?

I have a skill that— skills are what— I don't know how user-friendly they are for non-coders, but I—

so skill is something that an AI does. You could train it to do something, right? Like take multiple steps. And my skill was like out of control and stupid and like it did nothing for me and messed me up. So I think you're right. Maybe. It's not so good for non-coders.

Yeah.

Not yet.

You kind of gotta go over it and be extra detailed with the instructions. Coders are used to providing an uncomfortable level of like detail that really proofs any kind of, uh, mess-ups from a computer. Writing instructions to AI is kind of like writing code in that way. Writing code, you really have to make sure that everything is exactly clear. And writing instructions to AI, you also have to do that. But I have a skill and it doesn't work yet. Right.

Oh, really?

Yeah. No, I, I like to organize my computer in a certain way and conveying that has taken more time than, than actually organizing myself.

On the one hand, it's kind of magical. You can vibe code and talk to it and tell it to make stuff. On the other hand, you could get sucked into the weeds with that too. So I've spent hours and sometimes come up with nothing.

Yeah, no, I think it's a common occurrence that you dream of automating everything and then you realize that exactly specifying what you want takes a lot of time sometimes, and it's the debugging effort.

So if you were gonna tell somebody who's like a complete novice who just heard about vibe coding from this episode today, what would you tell them to experiment with? What would be the top 3 things you would do?

I don't know.

When you first download Claude or AI ChatGPT or—

Yeah, I think lovable is a good, a starting point. When I've done vibe coding workshops, I generally start there.

Mm-hmm.

But the first step is choosing something that you're excited about that could fit in a web app. So even if it can't fit in a web app, you can make a marketing site. You can, which is kind of like your pitch deck of why you're excited about this. And then you can go into Claude or ChatGPT, like you said, and kind of ideate with it. And then once you have a plan for, for a website that you wanna build, you paste it into Lovable out of ChatGPT or Claude, and then you just wait a little bit and it makes a nice little gratifying sound when it finishes generating. And then you have a website, which I think is like the new version of like a pamphlet or a pitch deck. And then yeah, it's, it's really fun in the end.

So when you say web app, so for people who don't know, like it's a website app. So for now, most of the stuff we vibe code would be a web app, meaning you go to a website like www.lovable.com or whatever your web that you built is like mine is under my website, mindbodyspace.com, sleep tips. The web app lives on my website and it doesn't go on the phone unless you go to my website on the phone, which is not the same thing as an app that you download from the App Store. Yeah, but Lovable, you can deploy it if I remember correctly. And be able to get an app that you could possibly put on the App Store, correct?

Yeah.

Yeah.

You can definitely make the transition. I haven't done that myself, but I, I think some people have done it fairly easily. I think it's more of a learning curve. I generally start with Swift and the traditional way of making an app. I haven't learned app building since all this stuff came out, so I'm still doing it the traditional way, except accelerated with AI.

So, Gabe, before we go, tell us a little more about Mindful Makers. What prompted you to start meditating? Why it matters so much to you? And I personally think it's awesome that you do this on such a regular basis. And tell us more about Mindful Makers in general and what happens and how people can join.

So Mindful Makers is a community of people who gather and build technology dedicated to promoting well-being in the daily lives of our community. So promoting well-being for me is heavily tied up with mindfulness and meditation. We meditate at the beginning to try to distill what it is that really promotes our well-being and sense of connection to others. Part of that is that I think people don't always know what they want. If they're not really in tune with themselves, then they easily will start accepting values imposed upon them from big corporations They might start scrolling on Instagram and they start identifying with the content they see there. Meditating at the beginning really helps us quiet the mind, the part of our mind that says, I would only be happier if I could send 10 more emails per day using AI, you know, or if I could, you know, vibe code the perfect website. A lot of the times technology is not necessarily needed for our wellbeing, but it does have a huge potential in some areas. So it's not always easy to know which are the areas where technology helps versus hurts in our daily lives. So it requires a lot of reflection, it requires a lot of experimentation. It's a lot of fun to do this. For instance, we have debates over whether if you really want to practice mindful eating, which is the meditation tradition that combines meditation with eating and apparently makes your food taste really good and makes you very grateful for it every time you eat. But does this have to do with technology? Maybe it does, maybe it doesn't. Would you really want to get out your phone every time you sit down to do mindful eating? I wouldn't. Some people there swear that they would. So it's a learning experience. It's a lot of fun. We're developing mindful technology and some of it helps some people, some of it doesn't. But I think there's a lot of cross-cutting benefit to some of it. For instance, technology that blocks harmful media or content or addictive social media is generally across the board stuff that people want. We build analyses using AI of, of the ideas that people have over what technology would count as mindful technology or promote their wellbeing. And a big part of it is the fact that they want to block other technology that's part of their digital environment.

Uh-huh.

And part of it is the need for greater self-reflection and self-knowledge as well. So those are the two themes that generally come up a lot. Along with exploration of something that has spiritual significance to them. It's oftentimes a, a book or, or maybe it's even a tradition or a ritual that's a part of their day. So these general 3 areas are the biggest areas of technology that people work on at these hackathons and share with each other.

That's awesome. And how did you get into mindfulness and why? Why does it mean so much to you?

I got into mindfulness in 2016. I discovered it through Headspace. At the time, there was an outbreak of mumps actually at Harvard, and I got it. And people on my track and field team were having to quarantine and wear masks before COVID And it was a big deal. The city of Newton shut down all their drinking fountains when I went home for the summer break. Yeah, because I was an early mumps patient in this outbreak. So during that quarantine time, I had nothing to do. I tried the Headspace app and it didn't actually stick with me the first time. Mm-hmm. Um, not until later. I was at high altitude training camp with my friends. Mm-hmm. And that's when I sat down to try it again because I had heard it was good for you. And at that time, Headspace actually was designed for athletes, or that's what I had heard. And there was lots of athletic stuff on there. So I thought it would make me a faster runner too, which is, you know, sometimes people start meditating and then they stick with it because it gives them so much more because they've heard it will make them a better writer, musician, and that's kind of what it was for me as well. I was bored and maybe anxious about being sick at first, and then later on I, you know, wanted to be a better runner. And now I do it because it's been part of my daily routine and it's part of taking care of myself, and I've observed how it's affected me in a bunch of different areas of my life. And so Mindful Makers is part of this project to share this, share meditation, but also try to add to it the technology piece because it's what I know. It's kind of like the author of Dilbert saying that he's not funny enough to be a comedian and he's not a good enough artist, but he does know a little bit about office culture. So he combines the three and out comes Dilbert. So I'm combining my knowledge of AI and rapid prototyping with my love for meditation. And I think there's a lot to do here and I think there's a lot of impact to be had.

That is a cool story. Your mindfulness actually started with technology then, right? Started on an app, right? I didn't really know that people just learned how to meditate on an app. But then you did other things, right? You go to the Insight Meditation Center in Cambridge now, right? Like a live community.

I started meditating in 2016. It wasn't until, I think, 2022 that I started with Cambridge Insight Meditation Center. So I think my journey was first the app. I had heard about the app. I wanted to be a faster runner and better at school and more chill. And so I, I stayed with Headspace for a long time. They gamified it at that time too. You could keep track of your, your daily streak, how many days in a row you had meditated for. And I, I remember I got up to 400 days and then I was so curious what it felt like not to meditate that I just stopped for a while.

And what happened when you stopped?

I think definitely like just a little more emotional turbulence was injected to, to my day. And that is, is fun at first. And then I thought, you know what? I like the stability. I like the calm. Mm-hmm. I like the control of that. I think meditation may not be for everyone, but I think it's for a lot of people.

I went to a talk by a Tibetan monk this week, and again, the idea of meditation is really just understanding and seeing your own thoughts. Really. And so the Tibetan word gom, G-O-M, for meditation is really just about getting familiar with your thoughts. So it's basically metacognition, which humans have the ability to understand and see their own thoughts in real time. Yeah, that's one of the reasons why we're a little more intelligent. Yeah, sometimes.

Yeah.

So I think it's actually an innate process. Yes. For our brains. And it's just a matter of how you use it. I love that you started with tech. That's interesting to me. But then you did go on to the live community and now you are combining also tech and a live community at your hackathons.

Oh yeah. These are live events in, in Boston at MIT. Yeah.

How old do you have to be to come to the hackathon?

I think maybe college age is— okay. We've, we've had some high schoolers too.

Yeah, but we talked about getting it out to younger kids.

Yes, right.

And you were developing a kit that perhaps we can bring to middle school or high school STEM communities.

Yeah, to learn meditation and to learn about how technology affects well-being very early. It's something that could have helped me even back when Instagram wasn't as addictive as it is now.

I call it Instagram Derangement Syndrome.

Yeah.

When you post something and you keep seeing who's DMing you. But this was such a fascinating conversation. Thank you so much for being here, Gabe.

Thank you.

I'm so happy that we met and I can't wait to come to another hackathon.

This is on the 30th, if you can make it.

Yeah. Keep us posted on your developments.

Yes, I will. Thank you so much for having me.

Thank you. That was Gabe Montague, founder of Mindful Makers. You can reach out to him on LinkedIn or head on over to his website at mindfulmakers. XYZ. Head on over to mindbodyspace.com/store. We have a limited offer for Relax to Soar planner. It has all the microhabits you need to thrive, and you get a free special session with me to get a head start on your goals and to keep you accountable. This planner has everything that I've been teaching for over 10 years at places like the Juilliard School, Columbia Law School, NYU Medical School, as well as to thousands of online students. On the website, you can also download your free PDF. Thanks for listening. I'll see you next time. This is Dr. Juna. You and your family wellness.