From journalist to iOS developer: How LinkedIn’s editor builds with Claude Code | Daniel Roth

Daniel Roth, editor in chief at LinkedIn, went from business writer to iOS app developer, without ever learning how to code. Using Claude Code, Daniel built and shipped multiple production-ready iOS apps to the App Store, including Commutely, a personalized train-tracking app for New York commuters. What you’ll learn: How to set up a dual-agent Claude Code system (builder + reviewer)Why being a “picky customer” is the right mindset for non-technical buildersHow Daniel prioritizes features using AI-ranked impact vs. build timeWhy saving everything as Markdown files creates long-term contextThe importance of branch-based development—even when AI writes the codeHow Daniel ships to the App Store without formal engineering experienceHis end-of-day “What did I drop the ball on?” Copilot workflow— Brought to you by: WorkOS https://workos.com/?utm_source=lennys_howiai&utm_medium=podcast&utm_campaign=q22025—Make your app enterprise-ready today Vanta https://www.vanta.com/howiai—Automate compliance and simplify security — In this episode, we cover: (00:00) Introduction to Daniel Roth (02:46) Daniel’s AI development workflow overview (05:56) Using Claude to prioritize feature ideas (08:58) Building vs. marketing (09:47) Creating a retention plan for his app (10:38) Introducing Bob the Builder and Ray the Reviewer (13:50) How Bob and Ray work together to build features (14:37) Why Daniel focuses on learning the process (16:34) The importance of using branches for development (17:39) Managing AI agents like managing a team (21:12) Navigating the App Store (23:06) Being a “picky customer” rather than a PM (25:00) Testing in Xcode and shipping to the App Store (28:14) Quick recap (30:00) Creating terminal aliases with Claude (31:38) Demo of his Commutely app (32:10) Using Copilot to manage work responsibilities (35:05) How Daniel talks to AI without personifying it — Tools referenced: • Claude: https://claude.ai/ • Claude Code: https://claude.ai/code • Cursor: https://cursor.sh/ • Xcode: https://developer.apple.com/xcode/ • Canva: https://www.canva.com/ • Microsoft Copilot: https://copilot.microsoft.com/ • Terminal: https://support.apple.com/guide/terminal/welcome/mac • Obsidian: https://obsidian.md/ — Other reference: • Commutely (iOS app): https://apps.apple.com/us/app/commutely/id6755789873 — Where to find Daniel Roth: LinkedIn: https://www.linkedin.com/in/danielroth1/ Newsletter: https://www.linkedin.com/newsletters/forward-deployed-editor-7378272989982683137/ — Where to find Claire Vo: ChatPRD: https://www.chatprd.ai/ Website: https://clairevo.com/ LinkedIn: https://www.linkedin.com/in/clairevo/ X: https://x.com/clairevo — Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email [email protected].

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Claire Vo?

  So you have actually named your clods and then you give them instructions to listen to each other.

• 0:06Speaker 01

  Yes, they have to talk to each other. One is called, the builder app is called Bob, Bob the Builder. And he's got instructions to stop constantly and you have to run everything by Ray, who's the review agent. Ray's job is senior software engineer who is obsessed with security. You have use code milestones, guard security architecture. And then the third agent is me. I am the person who breaks the tie that often happens between Bob wanting to do something and Ray saying you can't do it.

• 0:31Claire Vo?

  And you are doing all of this as someone who has not spent their career as a software engineer.

• 0:36Speaker 01

  For a while, I thought I'm like a mediocre PM. And then I was like, no, maybe I'm more like an architect. And now I realize like an architect actually knows real details. A PM is like super rigid and it keeps the entire app in their head and they're able to really prioritize well. I'm a bad prioritizer. All I am is a really picky customer. So I think that is like the role of the vibe coder is what do I care about deeply? I'm like walking through this house and I'm telling the architect, no, I want this room blue. I know you don't think it's a good idea. I'm telling you this is what I want.

• 1:07Claire Vo?

  Welcome back to How I AI. I'm Claire Voe, product leader and AI obsessive here on a mission to help you build better with these new tools. Today I had Dan Roth, who's the editor of LinkedIn and started his career as a business writer and editor, but has somehow become a software engineer, vibe coding iOS apps all the way to the App Store. He's going to show us his dueling Claude code setup that lets Bob and Ray build high-quality, production-grade software. Let's get to it. This episode is brought to you by WorkOS. AI has already changed how we work. Tools are helping teams write better code, analyze customer data, and even handle support tickets automatically. But there's a catch.

• 1:50Claire Vo?

  These tools only work well when they have deep access to company systems. Your copilot needs to see your entire code base. Your chatbot needs to search across internal docs. And for enterprise buyers, that raises serious security concerns. That's why these apps face intense IT scrutiny from day one. To pass, they need secure authentication, access controls, audit logs, the whole suite of enterprise features. Building all that from scratch, it's a massive lift. That's where WorkOS comes in. WorkOS gives you drop-in APIs for enterprise features so your app can become enterprise-ready and scale up market faster. Think of it like Stripe for enterprise features.

• 2:32Claire Vo?

  OpenAI, Perplexity, and Cursor are already using WorkOS to move faster and meet enterprise demands. Join them and hundreds of other industry leaders at workos.com. Start building today. Dan, welcome to How I AI. What I like about having people like me, maybe, that have a couple of years of experience under their belt is we've seen technology shift over time, and we've seen market shift over time, and we've seen our own career shift over time. And what I love about your story is you've seen a shift like this in an industry that you've been in, and you've seen how disruptive and scary it can feel. And now you're seeing this again, and you're really leaning into it.

• 3:17Claire Vo?

  So tell us how you came from your background, which is non-technical, into now we're going to see some power Claude code tips and iOS apps, kind of vibe-coded all the way to the App Store.

• 3:29Speaker 01

  I was a longtime business writer. I was at Fortune, Wired, Forbes. And I remember being at Fortune during the introduction of blogging. And all of the sudden, everyone had access to something that I thought was something that was unique to writers. We were able to talk to the world in a way that no one else could. And that was kind of like a monopoly, almost, on this ability to get ideas out into the world. And then WordPress came around, and Tumblr, and suddenly everyone could talk to everyone. And you discovered this explosion of ideas. And I have, and it was scary at the time, and then it was, you realize, oh, actually, you want to embrace this, more ideas, the better.

• 4:12Speaker 01

  And so you, as a writer, it changed, and as an editor, it changed how I thought about who I could go after and open up. Like, suddenly I realized there were way more people that could be sharing ideas and reporting than I ever thought were possible. A couple of years ago, when generative AI started making its way into coding, and it became clear that any of us could build anything, I had the exact same idea, but it went from being this, I was on the opposite side of it. It was like, oh, my God, I've had all these ideas of stuff I wanted to launch. And in the past, I would have had to go convince an engineer or a PM to go team up with me on something. And I was just trying to influence what got built. And then I could suddenly start building.

• 4:55Speaker 01

  So it has become an obsession of mine, of building to the point now where when I'm with friends and family, they're like, please stop talking about the stuff that you're building. You know, we don't want to hear about it anymore. And, uh, and I, I just have to like ratchet it back. But on the weekends, it is like, that is what I love spending my time doing. So like, while the Super Bowl was going on, I will be on my laptop, building either, like working on some new feature for the apps that I have out there.

• 5:18Claire Vo?

  Well, what's really funny is you are not alone. We just had Guillermo, the CEO of Versal on, and he calls that request for prioritization, like petitioning the government, where anytime you had something, you had to go to the government, which were unfortunately run by product managers and say, please, oh, please, government, would you pass this resolution to ship my button in

• 5:38Speaker 01

  Totally.

• 5:39Claire Vo?

  in production. And then we're also hearing from everybody, whether technical or not, that this is really fun to build. And it's kind of new hobby, new, new space for learning. And unfortunately, maybe, or fortunately, where they're spending too much time.

• 5:55Speaker 01

  Yeah.

• 5:55Claire Vo?

  So well, what we're going to benefit from your obsession now, and you're going to show us how you, as somebody who has a non-technical background, has used Claude code and some other tools to build real production apps. So what do you do? What's, what's your special sauce?

• 6:09Speaker 01

  Sure. So first, I should tell you, I started this with a course I took on Cursor. So it started with how to use Cursor, and I can share that link with you on the, it's a free course, how to build with Cursor, how to build apps. And I watched that, and so I started with Cursor. And then over time, I stopped using Cursor because I didn't really care where the files were kept. I wasn't going in and editing anything. I didn't actually need that. I just needed to tell people what I was doing. I needed to tell Claude what I was doing. So Claude code was the big unlock for me, and I'll show you what, how I build.

• 6:41Speaker 01

  So if you can see my screen here, I'm going to show you one thing I do is I keep, I use Claude and I ratchet back and forth between the $20 a month plan and the $100 a month plan, depending on where I am in my building. So right now, I'm in the $100 a month plan. I keep a Claude window up that has all of my, I keep everything in projects. So this is my Commutley project, if you can see this.

• 7:01Claire Vo?

  Yep.

• 7:01Speaker 01

  And I keep one running feature here, which is Commutley feature idea and tracker. So this is as people talk to me about what they wish the app did. And this is an app for being able to basically never run for the New York City train. Again, you can know if it's, I keep missing the train. I'm like, I'm going to build an app that is just for me, that is basically like, is the train almost here? Can I walk or do I have to run? That is what it's designed to do. It was that perfect product market fit because I was the entire product.

• 7:28Claire Vo?

  We call this personalized, so the rise of personalized software right now.

• 7:32Speaker 01

  Yeah, it's amazing. It's the best. And then suddenly you discover other people care about this also. And it's like, whoa, I have a community. So this is for my community of train runners. They give me lots of feedback and I've been keeping a, I keep, I keep one Claude chat available here that is just all the feature ideas. And I've given it a prompt to basically rank it in terms of time it'll think to build and the growth that it will think that this feature should have for the, for the app. And so the, I'll read you the prompt at the top, which is basically, let's use this as a running idea of ideas for Commutley as I log them, keep track of them and offer guidance, time estimate to build and estimated back and forth hours, potential impact score on

• 8:15Speaker 01

  two, one to three scales, customer happiness and growth impact. And then I've given it a bunch of ideas and I just keep feeding it new ideas. And it has this table here. So these are, if you can see the Commutley features ranked by build time. And so what I do is when I have free time, I go into this chat and I find a feature that I'm like, oh, I've got a couple of hours, I can go and build something and I'll look for it. So the things on my list to do are Siri integration, standalone widget. I've already built this SEO friendly blog, location awareness, but I thought maybe for this, it would be fun to build something I've been trying to do, which I built this scheduled automatic updates, which is, I know that every day at 7:30 a.m., I usually leave my house. I just want Commutley to tell me where the trains are. So I've built that, but no one's using it.

• 9:00Speaker 01

  And so I need to be able to build. I now have to think about discovery. By the way, this is the other part of this I've learned is like, The building is one thing, the marketing is a total other part of it. So now I've gone from being a quasi crappy PM to now being a quasi crappy PMM as I learned how to basically do marketing for my app too.

• 9:21Claire Vo?

  Here's, here's a funny thing that I'm, I'm thinking as you say that is I know I'm a true product founder and that I will literally build everything on this list before I can will myself to do anything that is like even remotely called marketing. And so you can get over your, uh, what we say in the industry, you know, if you can climb cringe mountain and just become a marketer, you will be very happy with the distribution of your app.

• 9:47Speaker 01

  So I've already worked with Claude Code on coming up with a retention plan. And I save everything as .md as markdown files. So I've got it within my commutely project, there are just, there's a document folder and then there's a list of markdown files. And I just, every time I'm working with commutely, every time we're working with Claude, I'm saying, write it into a file, log everything, log everything. And I do that for two reasons. One is the context window, Claude is constantly forgetting what it's working on. And then I'm forgetting what I'm working on because they only do this on weekends. So there, on Saturday I'll pick it up and I'll be like, what, wait, what am I building? I can't remember and how far did I get? So everything gets logged. That is one tip I try to give anyone that is in my, my shoes also.

• 10:32Speaker 01

  So we make these MD files. So this is retention plan and it's come up with a general plan, but now I need to go build it. So I keep two tabs, I use terminal and I keep two tabs open. One is for building and one is for reviewing. And now I've given Claude two personalities, which I can bring up by using, one is called, the builder app is called Bob, Bob the builder. And so I bring up Bob and this is my builder app and he's got instructions. Tell me your instructions on how you work. So this is just to show you what Bob is trying to do.

• 11:09Claire Vo?

  And quick question for folks that don't know how to do this. First of all, I have never seen somebody customize their Claude theme to let them know what version, um, flavor of Claude they have running. I've seen a lot of people that run multiple Clods and multiple flavors.

• 11:22Speaker 01

  Yeah.

• 11:23Claire Vo?

  Um, so I've never seen that. Pro tip, you, you can do that. Second is how do you, have you built these as aliases that have different settings? How do you build, actually, how do you technically make Bob? If you could just give us a high level.

• 11:35Speaker 01

  This is not going to be super satisfying because what I did was I go into Claude and say, "I don't want to have to tell you every single time what you are." So this is your requirement is Bob, and let me pull up what Bob is here. So, well, here, this is what I've taught Bob to do. This is basically my prompt on what I want Bob to do. So this is Bob telling me, Bob's idea, everything has to plan first. I've learned over time, like you don't build until it's done planning. And I've heard guests of yours say the same thing, like, "Plan, plan, plan first." And then, everything has to be built in modules. I don't want like tons of spaghetti code. I learned that with my first project, the code gets unwieldy. So I say, "You're a lean builder," whatever that means. And then, document everything, we talked about that.

• 12:17Speaker 01

  This is the important part, is I say to Claude, I say to Bob, "You have to stop constantly and you have to run everything by Ray, who's the review agent." So let me get over to Ray. So this is Ray.

• 12:29Claire Vo?

  I love this so much while you pull this over. So you have actually named your Clods?

• 12:36Speaker 01

  Yes.

• 12:36Claire Vo?

  And then you give them instructions to listen to each other?

• 12:40Speaker 01

  Yes, they have to talk to each other. So Ray's job is Senior Software Engineer, who is obsessed with security and with making sure that we don't leave what are, any of our design guidance. So I've got a whole design document also, and then make sure. So here's what Ray does. Ray, what I do, I review Bob's plans before he builds. He reviews code milestones, guard security, architecture, and what he cares about. And this is what was in my prompt is member trust, security, architecture, integrity, and quality. And I say, and because Claude is always rubber stamping everything, I'm sure your viewers see this all the time, everything's genius, you're always perfect. I have to tell it, you have to say no to things.

• 13:23Speaker 01

  Like you're, and then I break the tie. So the third agent is me. I am the person who breaks the tie that often happens between Bob wanting to do something and Ray saying, "You can't do it."

• 13:34Claire Vo?

  Well, can I suggest two other agents? One is Amy, the PM, that says, "Sorry, Bob and Ray, that hasn't been prioritized yet." And then there's Joe, the AE, that sells it anyway, whether or not it's been built. And then you have the full team, ready to go. Okay, so Ray and Bob work together. So how does Bob invoke Ray?

• 13:55Speaker 01

  So Bob has to, so let's do this. Say, I'm gonna say, "We were going to build the retention plan."

• 14:05Speaker 01

  All right, so I use the app. It finds its retention plan, MD. And then I just say, because Bob knows what his role is, I say, get started.

• 14:14Speaker 01

  So he should now go into plan Bob, I shouldn't even use, I shouldn't personify these people. So it now goes into.

• 14:19Claire Vo?

  Personify AI. I think it's very, it's, you know, it's just the way we work.

• 14:24Speaker 01

  Really? I feel so creepy about it. I hate it. I hate, I hate when I do it. So it's starting to read this markdown file that I showed you earlier of what this retention plan is. I came up with this weeks ago, reviewed it, I thought it was pretty good, and now we got to start building it.

• 14:37Claire Vo?

  This is amazing. I think this is so fun.

• 14:40Speaker 01

  So once it's done here, it will spit out something that I have to then copy and paste into Ray. Now, what I've done in the past is I haven't built Ray, and I've just said, run this past a security agent. That works, but what I don't like about it is I can't see what the security agent is saying. So what, what the builder agent will often just say is, security passed it, or, security wants me to change one thing. But because I'm trying to learn this also, I want to know what it's stumbling on.

• 15:06Claire Vo?

  I think that's an underappreciated workflow, which is, you know, earlier when you said, I got out of cursor because I'm not looking at files and I'm not editing things, the software engineer in me just started, like my ulcer starts to just flare. I am an aggressive code reader, but you are taking this. And I even think some of these like micro fictions that aren't necessarily like, yeah, could you spin up a Ray subagent and do all sorts of good stuff? But I do think putting a little bit of friction in the process where you're actually forced to like, copy and paste, put it over, read what Ray says.

• 15:38Speaker 01

  Exactly.

• 15:39Claire Vo?

  Does help with learning. And so I don't want people to, especially people that are moving from non-technical into more technical tasks, to make it so efficient that you're not actually learning the process, because then, like you, you can set up even more powerful systems.

• 15:54Speaker 01

  Couldn't agree with that more. That has been, and I'm not doing this just to build, I'm doing this to learn, but what I figured out early is that I'm not going to learn how to code. Like, just being realistic, like, I tried to do that, I got, I did find it that interesting, and I'm like, I'm a total beginner. And I'm like, well, it's actually building code. I just want to understand why it's building what it's doing. And the problems that I find that I can solve uniquely as a human are things of like, should we prioritize this or not? Or you're, this seems like it's going to be really expensive. Like, is this worth doing? Is this going to break my budget? My budget for Commutley is zero, except what I'm paying for Claude. And so I don't, I have to tell it all the time. Like, that seems expensive. I don't want to do this. And so here, this is, it's now done here.

• 16:38Speaker 01

  Let's come up with this, it's pretty long, implementation plan. So normally I would sit here and read this whole thing, but I told it, build it. Everything that Bob does has to be done in a, in a branch. That's one lesson I've learned. I used to ship everything to main and I, who heard the pain of that early on?

• 16:57Claire Vo?

  I yelled at a friend recently. I was like, make a branch, dude. Like, you're stressing me out.

• 17:03Speaker 01

  Oh, my God. But what I, I didn't realize this, and I found this out from real engineers later, is it's not, when I merged the branch, my other app, I merged the branch into main and it didn't work. There were, for some reason, it didn't like merge perfectly. It took me weeks to be able to figure out why it wasn't merging. So I never, I didn't, this was a lesson to me, is it doesn't always just work seamlessly. You probably know that well.

• 17:24Claire Vo?

  Yeah.

• 17:25Speaker 01

  Okay. All right. So here I've copying the plan here. Normally I would read this whole plan, but I'm not going to this time because I'm just going to build this into a branch and then work on it later. So Ray is now ready for this. I paste it in. There we go. And now Ray's going to get to work.

• 17:41Claire Vo?

  And so what Ray is doing here is the sort of, is this architecturally correct? Is it secure? Are there any, you know, as we say, there be dragons? Like, is there anything in there that you need to be worried about that Bob mixed? And I really like this. We, we haven't actually seen somebody do a persona to persona handoff. What we have seen is a lot of people will maybe use like an Opus 4.5 or an Anthropic model to build the plan. And then they'll actually hand it to like a Codex model. To review it, Codex is like a little, a little mean senior staff engineer.

• 18:14Speaker 01

  Totally.

• 18:15Claire Vo?

  And so it is, it is kind of nice to have that handoff and, you know, what I say is like dueling agents to just get, very similar to when you're working in a team. I think you can appreciate this, right? You write code and you're, or you write up, in my experience in product, like you write a product idea and then you bring somebody else into it and they're like, oh, but you forgot we have to have this compliance thing. Or don't you remember that when we did the data analysis on this, customers hated that piece? Or design says, actually, this is really technically hard to implement in our design system. And so I do think just taking some, taking some flows from your organizational process and then figuring out how to make them agentic is a really natural way for people who especially have been managers.

• 18:57Speaker 01

  Yep.

• 18:58Claire Vo?

  To start building out their own little AI productivity stack.

• 19:01Speaker 01

  It's such a great call. That manager, that idea of being a manager for this, you are managing these agents.

• 19:06Claire Vo?

  Yep.

• 19:07Speaker 01

  So like one of the things I find a lot with Commutely is it depends on, this app is really built around live activity on iOS. That is where you're tracking, you're trained, is your phone is locked, you just go and look at it. Live activity has, it turns out to have all kinds of limitations. And Claude will forget those limitations all the time. And so it will suggest things like, oh, you should just do this. And I'm like, we know, we've learned this already. Don't you remember? Like two weeks ago, we tried to build it. It turns out that like, it can't actually ping the MTA's API when it's, when, when the, the app is not in, on, on the screen. And so I, it's like being a manager of, I've read this somewhere else. This is not my thing, but someone once said that

• 19:48Speaker 01

  managing AI is almost like managing a really smart but hungover intern. And I feel that way all the time. It's like, you're a genius, but you don't have it this morning. Just remember, we've gone over this already.

• 20:01Claire Vo?

  Well, and AI is very similar to how I've positioned the benefits of what we call early career talent.

• 20:07Speaker 01

  Yeah.

• 20:07Claire Vo?

  Which is like, very capable, very excited, too inexperienced to know they shouldn't do something. So occasionally you get greatness.

• 20:15Speaker 01

  Right.

• 20:16Claire Vo?

  But we should use a branch and a PR just in case.

• 20:19Speaker 01

  That's exactly it.

• 20:20Claire Vo?

  As an AI founder, you're used to sprinting towards product market fit, your next round, or that first enterprise contract. But speed isn't enough for AI startups. Buyers expect security, compliance, and transparency from day one. That's why serious AI startups use Vanta. With deep integrations and automated workflows built for fast-moving AI teams, Vanta gets you audit-ready fast and keeps you secure with continuous monitoring as your models, infra, and customers evolve. AI innovators like Langchain, Ryder, and Cursor scaled faster and closed bigger deals by getting security right early with Vanta.

• 21:04Claire Vo?

  Listeners can claim a special offer of $1,000 off Vanta at vanta.com/HowIAI. Great, so now Bob has raised some edge case concerns, which I think are really

• 21:17Speaker 01

  Ray has raised the edge case concerns.

• 21:19Claire Vo?

  Yep, Ray has, and then says Bob's plan is solid. So then do you pass this back to Bob?

• 21:26Speaker 01

  Yeah, exactly. So now I copy and paste this right back.

• 21:30Speaker 01

  It's got the green light, and then Bob's gonna go build. Now, the next step in this is that I will then take this and put it into Xcode and do a new build, and then test it out in the simulator, see if it works, test it on my phone, and then I usually ship it to Apple right then. So I've been trying to do weekly updates to Apple. I will say the hardest part of building my two apps has been the App Store.

• 21:53Speaker 01

  I had no idea. Navigating the App Store is a whole separate. I've got so many chats where I'm like, what does Apple want from me here? And just getting Claude to teach me how to use the App Store has been a real... I feel like that's almost the last friction left in being a builder, is navigating the App Store.

• 22:10Claire Vo?

  I have a couple related requests for startups. We're all, we have, we have all the coding agents we need. They're gonna get better, we're, we're fine. We are spoiled for choice. We need AI for App Store submissions. We need AI for SAP Arriba submitting procurement requests and POs.

• 22:28Claire Vo?

  And we still need that perfect AI CRM. So all the builders out there, if you could give us these very boring but very high value problem solved, you have a lot of customers waiting out there.

• 22:40Speaker 01

  You know, one that I might build actually to solve, again, this idea of solving for your own problems is app store designs. So the, right now, the way that I do, you know, when you, when you, when you do something in the app store, you have to show what your app looks like. And I've been using Canva for it, but it's really, Canva's AI, I don't love, I find like I'm not getting what I need. So I might try to build an app that just takes people's screenshots and turns it into something for the app store.

• 23:05Claire Vo?

  I love it. Okay, so right before the plan, Bob is asking some design questions.

• 23:11Speaker 01

  Exactly. So this is great. So this is, I, I have, I've written about this in the past, but like, for a while, I thought, oh, I'm, I'm like a mediocre PM. And then I was like, no, maybe I'm more like an architect. And now I realize like an architect actually knows real details. A PM is like super rigid and like has understands, keeps the entire app in their head and they're able to really prioritize well. I'm a bad prioritizer. All I am is a really picky customer. So I think that is like the role of the vibe coder is what do I care about deeply? I'm like walking through this house and I'm telling the architect, no, I want this room blue. I know you don't think it's a good idea. I'm telling you, this is what I wanted. So the picky customer is, I think, the role, that is at least my role. That's how I think about my role.

• 23:54Speaker 01

  So in this case, Bob is saying, uh, there's some UX decision I have to decide.

• 23:58Claire Vo?

  I like this concept of a picky customer because I was going to say what I feel like is QA that's just looking at little edges and saying, no, that's wrong and that button isn't quite doing. So I like elevating this to, no, no, no, no, I'm not doing quality assurance.

• 24:15Speaker 01

  Yeah.

• 24:16Claire Vo?

  I am going through and being my own pickiest customer, you know? I'm, I'm sniffing my champagne before I drink it, as we say. And so that's a really fun, fun way to think about your role when you're building for yourself.

• 24:28Speaker 01

  Yeah, and it also does like my other app, which is a podcast clipping app. The, there was all this, it took a little while for the video to be made of the podcast clip and there was some kind of error. There was a message just saying, "Your video is being created." And I'm like, "Actually, I want this to have a better voice. I want this to be funny. I want it to say like, 'Oh my God, you clipped a great part of the podcast.'" And so as the picky customer, I could say, "This is the voice of this thing. Make it funny." And wrote some copy for that. So that's, that's how I think about it, more than QA, because you want to bring your personality into it.

• 24:59Claire Vo?

  Yep.

• 25:00Speaker 01

  All right. So in this case, it's asking me some options. It's giving me some options. I could pick my own option. Just for the sake of time, I'm going to just say, "Let's keep, let's keep the MTA blue." All right, and it's going to start building.

• 25:12Claire Vo?

  Great. And you're doing this all on the weekend. So are you doing, I have to ask, are you running multiple bobs and multiple rays? Do you single do it? What's your own, what is your own capacity for agentic management?

• 25:26Speaker 01

  Bob is allowed to spin off sub-bobs. So that is the one thing. What my, my prompt says, "You are allowed to have agents that you manage. You're responsible for these agents. I'm holding you accountable." And your sub-agents can't create their own sub-agents. So that's my rule on how to do this.

• 25:42Claire Vo?

  Got it. So Bob can spawn, but he's ultimately the directly responsible individual.

• 25:47Speaker 01

  Exactly.

• 25:47Claire Vo?

  Got it.

• 25:48Speaker 01

  Ray cannot spawn. Ray, Ray is Ray.

• 25:50Claire Vo?

  A Ray cannot spawn?

• 25:50Speaker 01

  No.

• 25:51Claire Vo?

  There's only one, one gatekeeper?

• 25:53Speaker 01

  Exactly.

• 25:53Claire Vo?

  And that gatekeeper is Ray and there shall be no more?

• 25:55Speaker 01

  That's right.

• 25:55Claire Vo?

  I love it.

• 25:56Speaker 01

  Exactly.

• 25:56Claire Vo?

  That's very principal engineer energy. It's like one guy.

• 26:00Speaker 01

  Yeah.

• 26:00Claire Vo?

  And you go to him and you're like, "Can we do this?"

• 26:03Speaker 01

  Exactly.

• 26:03Claire Vo?

  And only he can say yes or no. And he neither manages nor can be managed. That is, that is your Ray.

• 26:10Speaker 01

  Totally. Well, Claire, like one of the things that I have, I'm able to do this because I've now worked inside of a tech company for 15 years.

• 26:16Claire Vo?

  Yeah.

• 26:17Speaker 01

  And so I've watched how, so a lot of this is based on people I've worked on, worked with. And exactly to your point, the principal engineer, like I've watched my PM partners be like, "We should really run this past Sanjay." And you know, there's like one person who keeps coming up as like, "Just find out whether this is going to work or not, or where we're going to make a mistake." And so that is exactly what Ray is based on.

• 26:36Claire Vo?

  Every Ray that you've ever worked with is now like, is this, is this...

• 26:39Speaker 01

  Am I Ray?

• 26:41Claire Vo?

  Am I Ray? Okay. So we're watching Claude Code generate code as it does. And then, you know, maybe we won't show this part, but you're just going to take this into Xcode. Make sure it compiles, test it, and, you know, spend the rest of your life getting it to the App Store.

• 26:58Speaker 01

  Exactly. So let me just show you quickly, if I can, what this would look like. Is you go into Xcode. I've actually come to really like working inside Xcode, but I think I'm probably only using like 5% of it. I tried putting Claude into Xcode. They're like, they just built this new part of, where you can actually use Claude code in Xcode. I have not found it to be worthwhile. I kept running out of memory, so I just quit using it.

• 27:23Speaker 01

  But you go into Xcode, you do what's, you do Command Shift K, you do a clean of this.

• 27:31Claire Vo?

  Look at you, expert software engineer.

• 27:33Speaker 01

  I should say, I don't know what I'm doing. So let's just make that clear. I have like learned these things. I'm basically the bot. I'm like, Claude has told me what to do.

• 27:41Claire Vo?

  You're the agentic browser.

• 27:42Speaker 01

  Exactly. It's exactly it. Like, tell me how to do this. I'm doing it. I'm just like, someone's pulling my strings. And then you go to build, which is what, Control B. It's going to run this thing. When it's done, it'll pop up the simulator. I can test it out. The simulator is, I find, like 80% close. But you really like, there's so many edge cases you don't even realize until you get it onto your phone. I test it on my phone, and then I quickly ship it to Apple, get it in Test Flight, test it for like an hour, and then I just ship it to the App Store.

• 28:14Claire Vo?

  And then you have all this extra time so you can make it to your train on time.

• 28:19Speaker 01

  Exactly. That's exactly it.

• 28:21Claire Vo?

  This is an awesome flow. Just to recap for folks, so what we saw is sort of dueling Claude codes. One's a builder, one's sort of like an architect reviewer type. You build your, you have a standing prioritization roadmap chat in the Claude web or desktop app, where you're always putting in ideas, reprioritizing them against like a simple three-point framework. On the weekends, you pluck one of those, you make a PRD, you give Bob the PRD. Bob invokes Plan Mode in Claude code. It builds a plan. You copy and paste that over to Ray. Ray either greenlights it or gives feedback. You give it back to Bob. Bob builds. Love it.

• 29:04Claire Vo?

  And then you use keyboard commands you don't even know in Xcode, and you ship it to the App Store.

• 29:13Speaker 01

  That is 100% right.

• 29:15Claire Vo?

  And you are doing all of this as someone who has not spent their career as a software engineer.

• 29:21Speaker 01

  I've been around a lot of software engineers. I've watched them work, but no, I have never done any of this myself.

• 29:26Claire Vo?

  This is, I mean, this is a flow we really haven't seen. I think it's a useful, really repeatable one that others can use. And I, you know, if anybody in San Francisco used public transport anymore, then I would definitely request a regional option for us. But in the meantime, if you could build a Commutley that tells me like when I need to yell at my kids at what level to get out of the door to school on time?

• 29:54Speaker 01

  Yeah.

• 29:54Claire Vo?

  That would be, yeah, apparently.

• 29:56Speaker 01

  Apparently is next.

• 29:57Claire Vo?

  That's what we need. I'm gonna, I'm gonna replicate your flow and build that next.

• 30:00Speaker 01

  Nice. Alright, so here it is. Commutley is now opening on this, uh, oh, nope, I don't have Metro up. I can never remember the command.

• 30:09Claire Vo?

  I'm gonna have to suggest for anybody who can never remember the command.

• 30:13Speaker 01

  Yeah.

• 30:13Claire Vo?

  You should watch our episode with John Lindquist, who shows us how to set up terminal aliases. You can just ask Claude to set up a terminal alias. That will then run these like regular things that you need to run. So if you wanted to type in metro.

• 30:28Speaker 01

  Yeah.

• 30:28Claire Vo?

  And have the terminal run it, you could create a terminal alias using Claude code.

• 30:32Speaker 01

  That's so much better.

• 30:34Claire Vo?

  Yep.

• 30:34Speaker 01

  I have it like in a notes file somewhere, which I can then never find. So that's, that's much handier.

• 30:38Claire Vo?

  Let's go back to Claude code. I'm just, I gotta have you do this. We'll do it live. This can be a live HowAI and say, can you make me an alias for the metro command? So it's easier to run in the future?

• 30:51Speaker 01

  By the way, I've really tried to not say, can you or please? This is part of how I'm like, making sure that I don't personify this thing. This is how I stop the robots from taking over.

• 31:01Claire Vo?

  This is, this is one of two things. This will go to our last question, which is how do you talk to, talk to the bots? I am so southern. I'm like, would you pretty please do this? I, please, sir? Yes, ma'am. The other thing is actually, quick parenting tip. I always finally ask my kids, like, would you, do you want to unload the dishwasher? And they're like, no, I don't want to unload the dishwasher. And so that's another, another thing is you don't want your AI to say no. Okay. So now anytime you type in metro, it'll restart the bundler.

• 31:32Speaker 01

  Amazing. Thank you.

• 31:33Claire Vo?

  You're so welcome. HowAI, live.

• 31:36Speaker 01

  Wow. Just learned something new. I love that.

• 31:38Claire Vo?

  Okay, so let's see if it's running now.

• 31:40Speaker 01

  So let's open the simulator.

• 31:43Claire Vo?

  Look at it.

• 31:44Speaker 01

  There it is. So this is exactly what I wanted to build. This was a hidden feature on Commutly is that you could set it up to go off at certain times. Bob and Ray just worked together to build this thing. And now every day, 7:00 AM, weekdays, I can get this. But now if I go to my morning commute,

• 31:59Speaker 01

  it's going to show me these trains are ready. I get this. If I go to the lock screen, it'll show up here. So if you were at 14th Street, your next train's coming in five minutes, you gotta start running.

• 32:10Claire Vo?

  I love it. Okay, download this app on the App Store. Give Dan all the feedback so he can give it to Claude, Bob, and Ray. All right, Dan, let's hop to some Lightning Round questions and then I will get you back to your day job. So we get to the end of the week and on the weekend you can start playing with some more Claude code.

• 32:27Speaker 01

  Thank you.

• 32:28Claire Vo?

  So the first question I have for you is, are there any sort of non-coding workflows or tips or tricks that you think are really useful that people can pick up in like five minutes that you think are going to make their life better?

• 32:38Speaker 01

  We've talked this entire episode about something I just do on weekends. But most of my time is actually spent at work. I run a 400-person team. I've got a contact switch all day long. I rely on Copilot constantly to do that because it has access to all my files and my entire team. So my command that I start the day with or I end the day with usually is, what did I drop the ball on? And I'll show you that. What did I drop the ball on? And then I usually, and then for the point of this show, I'm going to say anonymize

• 33:12Speaker 01

  any names or project names as this will be seen by people I don't work with.

• 33:22Claire Vo?

  Perfect.

• 33:23Speaker 01

  Okay. It's going to go through Outlook, it's going to go through Teams, it's going to go through any updated files. It knows who I report to, it knows who reports to me, and it keeps track of things that I'm constantly clicking on during the day. So it'll find, it'll go through all my emails and find places where I'm not responding or teams where I'm not responding. And what's great is it's not just places where I've been app-mentioned. It's stuff that it knows that I am actually interested in. So projects that I've been like, kind of like following over time.

• 33:53Claire Vo?

  What I think is really useful about this is we've seen a lot of people do their morning daily digest, but we haven't seen anybody do their evening nightly nudge, which is, hey, you got through your whole day and you forgot X, Y, and Z. And how many of us, especially managers, want to start the morning with nothing to do and having wrapped it up all in the evening? So I think just even take AI out of it, moving that practice from upfront in your morning to in the afternoon, when you can actually do something about the stuff that ate your plate during the day is a really great idea.

• 34:24Speaker 01

  That's exactly it. For me, this is my 30 minutes before I leave. I ask this, I do this prompt and then I can go through here. So here's an inbound escalation from a long time creator. Someone's unhappy with me. I didn't reply. I have a draft reply that I never sent.

• 34:38Claire Vo?

  My gosh, this is my life.

• 34:42Speaker 01

  And this is it. This is like, you get at some point, you start managing so many different projects. You're like, I know I'm dropping the ball. Just tell me what I am dropping the ball on. So this is my savior. This is like what I count on AI to be able to do to make me better at my job so that I can spend the weekends building money losing apps like Commute Lee.

• 34:59Claire Vo?

  Yeah, you can vibe code code. You cannot vibe code gross margins. So

• 35:04Speaker 01

  That's true. That is true.

• 35:06Claire Vo?

  All right. And then my last question, I ask everybody, but when AI is not giving you what you want, you've said you don't want to personify AI, which I, I do. So we are on opposite sides of this. Very interesting. What is your, then what is your tactic? Do you, do you like kill Bob and his spawns? Like, do you, yeah, are you an all cats person? What's your prompting strategy?

• 35:28Speaker 01

  I mean, there are times where, so I'm usually really just straightforward. I just, I'm like, this is what you're doing. I try not to say please or thank you because again, I just don't want, you know, at some point when we get to AGI, they're going to be running the show and I just, I'm fighting until we get to that point. I'm holding on to my humanity. And then, but I usually just, I usually just spend a lot of time saying, we've gone over this already. You've talked about this. Go search your memory to go find the time where we've done this. And it is, it's a lot of parenting. I have, like you, I have three boys and I rely on a lot A lot of the things where I'm like, we've talked about this, remember? And, you know, and your kids are like, you know that they are, they want to do the right thing.

• 36:09Speaker 01

  And so it's just a lot of like helping them get there. There's something we say that I learned from a former manager and that we say a lot at LinkedIn, which is assume best intentions. And that is how, that's by the way, a big change from how newsrooms work, which was my last life, which was you assume worst intentions all the time. At, at tech companies, you assume best intentions. I've now taken that to my family and I've taken it also to building with AI. So I assume the AI has best intentions, but has to be reminded about how we work. So I don't yell. I'm pretty clear and I try to give it a little bit of grace.

• 36:41Claire Vo?

  Perfect. Well, Dan, this was awesome. I think it's a great overview for technical, non-technical, busy, busier alike. So thank you for sharing your flows with us and our audience. Where can we find you and how can we be helpful?

• 36:55Speaker 01

  So I'm on LinkedIn, not surprisingly. Find me, you can follow me on LinkedIn. It's Daniel Roth at LinkedIn. And I've got, I keep a running newsletter called Forward Deployed Editor, which tracks how I'm learning how to AI. And it talks, it's really designed for people who are non-technical to be able to get, to be able to build cool things. And so I try to talk about what I'm failing. It's mostly what I'm failing at, what I'm learning, what didn't work, how to avoid making the same mistakes I've made. And so if you can follow that, that'd be great. Otherwise, I'm always constantly posting on LinkedIn, great content that creators are posting around the world. So I hope you'll follow me there.

• 37:32Claire Vo?

  Great. Well, thanks for joining How I AI.

• 37:35Speaker 01

  Thanks for having me.

• 37:37Claire Vo?

  Thanks so much for watching. If you enjoyed the show, please like and subscribe here on YouTube, or even better, leave us a comment with your thoughts. You can also find this podcast on Apple Podcasts, Spotify, or your favorite podcast app. Please consider leaving us a rating and review, which will help others find the show. You can see all our episodes and learn more about the show at howiaipod.com. See you next time.