My extremely personal Vibe Code diary

by

Hello, fellow Vibe Coders. I am a new Vibe Coder who left the handmade shoe industry in Korea to embark on a startup journey. I’d like to share a record of the past two months or so.

As I am Korean, I am using Google Translate to generate the English text for each paragraph; I ask for your understanding.

1. In the two-plus months since I started using Claude Code, I have built nine simple apps; I released three of them on iOS and five on Android. This count includes projects I abandoned midway because they became too difficult.

2. The total number of downloads is just over 200, and I’ve generated one paid subscription and $15 in AdMob revenue. These are the results from June through July. I hope this post will serve as a record of my early, modest beginnings.

3. Making the app was easier than I thought. (However, I have great respect for developers. In fact, depth is more important than how quickly it is made; if you miss this, it cannot be called a proper app. I already know this from my 10 years of experience handcrafting dress shoes.)

On a personal note, planning and developing apps is a process of transforming delightful ideas into tangible, real-world assets. While I truly enjoy this work, making a living from it was a different challenge altogether—so much so that I began taking on part-time jobs at the age of 40. That is fine by me, but because I love this work, I aspire to transition into a professional career. In this context, being a "professional" means reaching a level where I can generate sufficient income from this field.

When I first started learning about ASO and app marketing, I was struck by their similarity to SEO: because everyone has access to the same information, the playing field eventually levels out. Anyone can use effective keywords, so without targeting a specific vertical, you are bound to fall behind. I call this the "entry phase." Like you, I successfully navigated this stage. However, the next challenge awaits: conversion.

Anyone with experience in e-commerce knows that the 4P strategy is used to boost conversions. Alternatively, one can keep the conversion rate constant (mine, for instance, is a very average 5%)and focus on maximizing impressions. External advertising and social media activities are examples of ways to maximize impressions.

I have recently begun testing in this area. You are likely familiar with terms like PoC, MVP, and PMF. Personally, I used to find these concepts rather vague and abstract. While some might be lucky enough to find the right fit quickly, others might not; skill certainly plays a role, but luck cannot be discounted either. With this perspective in mind, I am exploring ways to maximize exposure.

I viewed short-form platforms like YouTube and TikTok—which dominate my daily life—as virtually the only channels where content is accessible to everyone. That is why I created a "YouTube Lab." I will not mention or discuss my specific YouTube account here, as doing so would compromise my authenticity. I use a selfie and a sample of my own Korean voice to create a "chroma-key" narration layer that makes it appear as though I am actually speaking. Over this, I use Veo3(an open-source video generator)to produce visuals that dramatize the specific problem I aimed to address when developing this app. The goal is to foster viewer empathy; if viewers resonate with the content, they are more likely to keep watching. Finally, I record footage of the app's actual UI and edit it all together. That is the entire process.

Now, let’s get to the main point. I recently succeeded in automating this process, which has enabled rapid video production and simultaneous uploads across multiple platforms. In fact, downloads have grown by over 10%(a remarkable achievement accomplished just one week after I started making videos, acting like a breath of fresh air for my previously stagnant app portfolio)and I am now focusing on improving video quality. The problem this process addresses is that indie developers (who, like me, may fear appearing on camera or struggle with public speaking) can promote their apps without significant cost. It also allows them to quickly gauge, through video performance, whether consumers resonate with the problems they’ve identified and the solutions they’ve proposed. If this post draws interest, I’ll be back. I’m rooting for your work. From Seoul.

138 views

Add a comment

Replies

Best

This was genuinely inspiring to read. Your shoemaking comparison says a lot: AI can help you produce quickly, but depth and craft are still what make something worth keeping.

At this stage, I’d probably stop measuring progress by how many new apps you can ship and pick the one whose videos get the strongest response. Make 20–30 variations around that single problem, study which hooks bring downloads and comments, then use that feedback to improve the product itself. Your automated video workflow could become more than a promotion tool — it could be a fast way to test which problem is actually worth going deeper on.

 It is truly gratifying to know that there is someone who shares my perspective on the core essence of this work. I also view the app's promotional impact as merely a bonus. The reality is that even when developing an app using "Vibe Coding," a significant amount of time and effort is required to reach the deployment stage. If the initial problem identification is flawed, it results in a massive opportunity cost—especially for solo entrepreneurs or small businesses. My automated video testing process aims to drastically reduce this cost. I would be deeply grateful if you continued to follow my journey.

 Absolutely. Reducing the cost of testing the wrong problem is a huge advantage, especially for solo builders. I’ll definitely keep following your journey — really curious to see which ideas your process helps validate fastest.

Respect for sharing real numbers instead of "I made $10k in 30 days." Posts like this are much more useful for people just getting started.

 Thank you for taking the time to leave such a thoughtful reply to my post. As a newcomer just starting out, I realize there is a lot for me to learn. While I am reading plenty of books, I believe the most important thing is the ability to take quick action rather than getting stuck in the planning stage.

Your background probably gives you an advantage. spending ten years perfecting handmade shoes teaches patience, and that transfers well to building products.

 Thank you for your kind response. I understand that my experience may differ from others', and you have a very accurate grasp of how my experience influences my coding. I would appreciate it if you could continue to join me on this journey.

I think your YouTube strategy has potential because you are selling the problem before the product. Thats usually much more engaging than a traditional app demo.

 I strongly agree with your thoughts. While streamlining this process is important, I believe this video project holds even greater significance as a replacement for app demos. It could even open up new business opportunities. Please join me on this journey!

I am curious have you noticed whether YouTube Shorts, TikTok, or another platform is driving the best installs so far? that comparison would make a great follow up.

 Based on current data, YouTube is showing the highest number of views and downloads. Recognizing the importance of the platform, I’ve set up a simple tracking mechanism to monitor traffic sources; however, my ultimate goal isn't to become a famous creator, but rather to build an automated video system capable of replacing standard app demos. I hope you’ll join me on this journey!

 That's a smart way to think about it. If the videos can consistently outperform static app demos, that's a win regardless of follower count. Looking forward to seeing how the automation evolves. Best of luck!

Nine apps in two months is wild output. The video automation is the real find buried in here, most solo devs treat marketing as the thing they'll figure out later and never do.

 That is a point I am focusing on as well. Marketing requires a significant investment of both time and money. Like a software developer, my own nature—rooted in the craft of shoemaking—was focused on the act of creation itself. However, the true essence lies in "problem-solving"—something that can only be fully realized when people become aware of our product. I ask for your continued support as I move forward on this journey.

The number everyone here skipped is the one paid subscription. Nine apps, two hundred downloads, and one person actually reached for their wallet. That single person is the only real signal in the post. Everything else is noise at that sample size. So the question I would spend the next month on is not which platform drives installs. It is which app was it, who were they, and what were they doing in the ten minutes before they paid. You already understand this better than most people in this thread. A handmade shoe sells because of one specific foot. Ten years of learning that, then two months of shipping nine pairs and hoping a foot turns up. The craft transferred. The fitting has not, yet. Real respect for posting actual numbers instead of a screenshot of a good week. Which app was it?

 Thank you for your insightful comments; I resonate with what you’ve shared. To give you an example of my most successful service, I developed "CatchNote." It is a knowledge storage service that uses AI to analyze the context of URLs, PDFs, and images (via OCR)—regardless of the source—and automatically organizes them into directories for storage. The core issue I am focusing on is that modern people are exposed to an overwhelming amount of information that passes by all too quickly. Take "health information," for instance—something I find particularly useful; I gather it from various sources, but trying to retrieve it later feels like rummaging through a cluttered room. I believe that minimizing this friction will become increasingly important in the AI ​​era, and I am currently looking for others who share this perspective.