← 대시보드로
2026-05-19
2개 영상
2026-06-19 05:50 생성
개발자 왜 씀
AIMAX_PD 2026-05-19
개발자 왜 씀
↗ 유튜브에서 보기

핵심 요약

거대 플랫폼을 15분 만에 코딩 없이 복제한 제이슨의, 바이브 코딩으로 앱 만들어 돈 복사하는 3치트키.

주요 포인트

  • 클로드 디자인으로 살아 움직이는 첫인상(코덱스 시안→5분 웹)
  • 코덱스의 뇌로 백엔드를 평범한 영어로 구축(회원 관리·데이터 연동)
  • 구글 로그인·캘린더 API 연결+AI 자동 테스팅(1인 창업 연 19억 사례)
이 남자는 천재입니다. 연 매출 사조 원이 없는 거대 플랫폼을 단 15분 만에 코딩 한 줄 없이 똑같이 만들어낸 기업과 제이슨. 그가 공개한 바이브 코딩 하나로 나만의 앱을 만들고 돈을 복사하는 세 가지 지트키 저장하고 끝까지 보세요. 첫째, 클로드 디자인으로 압도적인 첫인상을 만드세요. 제이슨은 복잡한 디자인을 쓰지 않습니다. 코덱스로 뽑아낸 시안을 클로드 디자인에 던져 주기만 하면 모든 버튼이 살아 움직이는 고퀄리티 웹사이트가 단 5분 만에 완성됩니다. 둘째, 코덱스의 뇌를 빌려 백핸드를 단숨에 구축하세요. 가장 똑똑한 채치 최신 모델을 활용해 당신이 원하는 기능을 평범한 영어로 설명하기만 하면 됩니다. 복잡한 데이터 연동부터 회원 관리까지 AI가 알아서 코드를 짜고 실행해 주기 때문에 개발 지식이 전혀 없어도 누구나 사장이 될 수 있습니다. 셋째, 구글 연동과 자동 테스팅으로 비즈니스를 완성하세요. 구글 로그인과 캘린더 API를 단 몇 분 만에 연결하고 AI가 직접 브라우저를 움직여 오류를 찾아내게 만드세요. 실제 1인 창업자들이이 방식으로 연간 19억 원 이상의 수익을 올리고 있는만큼 검정된 시스템을 빠르게 구축하는 것이 핵심입니다.이 내용을 더 자세하게 정리했습니다. 자세한 내용은 본문을 확인해 주세요. 이런 바이브 코딩 수입화 이야기 재밌었다면이 채널을 구독해 주세요.
Your Whole Team Uses AI. Why Hasn't the Work Changed?
Dylan Davis 2026-05-19
Your Whole Team Uses AI. Why Hasn't the Work Changed?
↗ 유튜브에서 보기

핵심 요약

"AI를 쓴다"는 같은 말이라도 단순히 이메일을 다듬는 사람과 AI가 자율적으로 전사·후속·CRM 갱신까지 하는 사람은 완전히 다르다. AI assisted와 AI native를 가르는 4가지 질문으로 작업의 병목을 찾는 법을 설명한다.

주요 포인트

  • 잘못된 질문은 "AI를 얼마나 자주 쓰나"가 아니라 "이 작업에서 AI가 어디서 막히나"다. 그 병목을 풀면 레버리지가 커진다.
  • 1) AI가 볼 수 있나(can it see it): 필요한 컨텍스트에 접근하게 하되 AI가 소화하기 좋은 형식(텍스트·CSV·마크다운)으로 줘야 한다. 컨텍스트가 머리를 채우면 지능이 급락한다.
  • 2) AI가 이해할 수 있나(understand it): "보면 안다" 같은 주관적 표현 말고 binary(예/아니오) 기준으로 "좋은 결과"를 명시해야 AI가 스스로 채점하며 기대치에 다가간다(예: 담당자 명시, 마감 명시, 120단어 미만).
  • 3) AI가 실행할 수 있나(act on it): 대부분의 connector는 읽기만 가능하다. 이메일 초안 작성, CRM 갱신, 작업 추가처럼 쓰기까지 하려면 보통 Codex·Claude Co-work 같은 desktop agent가 필요하다.
  • 4) AI가 스스로 개선하나(close the loop): 가장 많이 건너뛰는 단계. AI가 교훈·선호·인사이트를 파일로 저장하면 작업이 매주·매달 가치가 복리로 쌓이는 자산이 된다. 회의 요약 예시처럼 녹취 자동 수집→이해→초안·CRM·작업 실행→승인·학습으로 이어진다.
Two people can both say I use AI every day and mean completely different things. One rewrites emails with ChatGPT, while the other has AI reading transcripts, drafting follow-ups, and updating their CRM all autonomously. Same words, different planets. If you're new here, I'm Dylan. I run an AI consultancy. And this gap comes up in almost every coaching session that I do. So here's a four-question test to find the exact spot where AI breaks down in your work. I'll show you also a live example that everyone gets wrong. So, let's get into it. You may hear a lot of people talk about being AI native, either personally or within an organization. And this term gets thrown around a lot, but very rarely is it lived up to. Often times when people talk about AI native, this sits on a spectrum, and most people they're AI assisted. Reason being is they're using AI on top of old ways of working, and they have not really even changed much. AI native individuals and companies, they fundamentally change the way they work, what they feed the AI, how much reliance they put into the AI in a variety of other things. So my goal of this video is to walk you through the four-question test to understand what activity sits where on that spectrum of being AI native versus AI assisted. And it's important to note that this scales both up and down. So being AI native could be you just inside of your job making sure that you're using AI to its best capabilities with the access you have, or it could be at a team level, it could be at a department level, or even an entire company level. Now obviously the complexity of being {quote} AI native increases as you go up. So it's obviously more difficult to be an AI native business than it is to be an AI native individual. So my recommendation for most people is simply starting with yourself, getting very good at utilizing AI in an AI native way, and then scaling out in complexity slowly to your team, your department, and then your business. Now one commonly asked question, which is completely the wrong question, is how often do I use AI? Some people equate that to being AI native. So if my team uses ChatGPT daily, that means they're more AI native than others, which is somewhat true, but again there's levels of being AI native. This is probably level zero or one. There's probably five or six levels. And the reason this is the wrong question is just because somebody uses AI doesn't mean that they're using it to its fullest capability for that given activity. Instead, the question you want to ask yourself is where does AI get stuck in my workflow for this activity? If I discover that, I can release those bottlenecks to get more leverage from AI for that activity and other activities as well. Quick pause in the regular programming. This video is brought to you by me, as always. Two quick things. First off, below is a 30-day AI Insight Series, completely free. You'll get 30 insights in your inbox so you can apply AI to your business and your work. The second thing is if you'd like to work with me, below are a series of offerings to see if there's a good fit between the two of us. Now, let's get back into the video. Now, what are the four questions? I'm going to give you a quick rundown here, but then we'll walk through each one in detail. So, the first question is can AI see it? What I mean by it? I'm referring to the context that AI needs to have access to to achieve a given task. That's the first question. The next one is can AI understand it? So, this is all around taking my perspective and giving it to AI so it can give me the same quality output that I would do myself manually. So, I need to get what's out of my head into the AI's head. After that, we have can AI act on it. Now, this comes down to AI's access to other tools. Is it able to actually write to those tools or can it only read to those tools or does it have access to tools at all? You may have to copy and paste information into ChatGPT instead of it connecting into your email, your CRM or whatever else. And then finally, this one's more long-term, but it's closing the loop. So, can AI improve on itself over time? Is it compounding in its lessons learned from how I like to interact in a given activity as well as the insights it's derived week over week, month over month? And this is often one a lot of people skip. So, those are our four questions. Now, let's jump into the first one in more detail, which is can AI see it? And this is foundational. If the AI can't see the context it needs to achieve a given task, the entire rest of the flow is a waste. And this is more than just giving AI access to the context, but it's also making sure that context is in a format that's suitable for AI because not all file formats are created equal. If I pass AI a video or really large PowerPoint, it's going to fill up its head really quickly and then if you fill up the head too much, it gets dumb. It falls off like a rocket's intelligence. And I've ta