Right here, I've got four different agents that are looping, calling other sub agents, and writing all these prompts for me, and designing systems for me. But, is this actually productive, or is that just a cool demo? Here's your monthly reminder that you shouldn't be prompting coding agents anymore. You should be designing loops that prompt your agents. Boris Cherny and Peter Steinberg publicly said they no longer prompt their coding agents. They write loops. A loop is three things: a trigger, an action, and a stop condition. If you're still writing loops that prompt coding agents, you're falling behind. You need to build a meta agent that infers what loops you would have wanted based on your vibe, and then write those loops. We're seeing a ton of talk about agent loops, loop engineering, whatever you want to call it. So, I wanted to make a video to clear up what that actually means. Because I think that everyone kind of has their own spin and a different definition of what this is, and it applies to everyone very, very differently. I think that this definition sums it up pretty well. Loop engineering is replacing yourself as the person who prompts the agent. You design the system that does that instead. A loop here can be thought of as a recursive goal, where you define a purpose, and the AI iterates until complete. And there's really two most important pillars of that in my mind, which are the goal. What is the actual objective? Something typically that's objective, not subjective. And then verification. How does the agent know what that stop condition is? How does it check and iterate? So, anyways, if you take all that advice, and then you start doing stuff like this, and designing swarms and fleets of agents that constantly run 24/7, then you need to think about what are you actually doing here? And is this actually moving the needle? So, first of all, I thought to myself, how do I actually use agent loops? Because when you read some of those tweets that I just showed earlier in this video, you kind of think to yourself, okay, if I'm not having five agents that are continuously around the clock orchestrating five of their own agents, then I'm falling behind, or I'm not using my cloud subscription in the best way. And I think that that's very false. Because if you don't understand what you're doing, then you're probably just going to scale problems, and you're going to have a ton of bugs and a ton of things that you're going to have to fix later. And also, not all of us are in a scenario where having agents work 24/7 around the clock actually benefits us. For example, I don't. I have agents that do things on a certain cadence and I have agents that do things based on certain event actions, but just having them do 24/7 work for me isn't helpful. I think if I was working with a team on a codebase and we were building a product and we were constantly iterating and pulling in different things, then it would maybe make more sense, but for me, that doesn't apply. So, I just wanted to come in here, explain this as simple as I can, and hopefully shed some light on where you guys can start applying loops into your workflows and why and how. So, I actually built an agent loop for this HTML that we're going to look at today, and it basically went through a ton of different sources. It checked 45 sources, whether that was articles, YouTube video transcripts, X posts. It looked through a ton of stuff, and then it kept looping on this until it had a good idea of what to build. And then once it built this HTML, this wasn't V1. This was probably V7. It had to keep checking, screenshotting, reviewing, iterating, and then it finally said, "Okay, we're done. This is what we got." So, let me walk through this with you guys. An agent loop is just an AI that reasons on what to do, acts on what to do, starts implementing, and then it observes the result. And it will do that over and over and over until some sort of goal is met, until it knows we've hit the stop criteria, this is good, I'm going to stop now. And a really simple visual that I like to think about is AI is never perfect, right? It's never going to one-shot something and you just accept that final output. And so, if we have attempts on the x-axis and we have quality on the y-axis, let's think about this. On attempt one, if you are just giving your agent some sort of simple task, maybe you get to like, let's just say attempt one, you get to 50%, and then you look at that and say, "Okay, here are some changes to make." And then by attempt two, maybe you bump up another five or 10%. And every time that you give more feedback and iterate, you just kind of keep moving up on quality until you hit somewhere where you're okay with that, 90, 95%. And so, the whole idea is why don't we outsource this part, this feedback and iteration loop, to an agent rather than having the human do that? Cuz this is going to happen either way. So, if we have an agent do that instead of a h

I've just built one AI system that does basically everything inside of my business and it replaced every other AI tool that I was already paying for. And no, this isn't some complicated coder thing. So, here's the problem it fixes. Most people have one tool for emails, another one for finding leads, another one for content. And you're just re-explaining your business every single [music] time. So, instead, I built one AI system that already knows my business and I just tell it what I need. So, watch this. I type, "Write this week's newsletter." and it writes the whole thing in my voice and then I could type in the same chat, "Find me 50 new leads in my area." and it's done. So, it figures out which job it is and it just handles it. And here's how you could build your own. Open up ChatGPT or Claude, start a new project and paste in everything about your business. So, what you sell, who you sell to, and how you talk to your customers. That is it. So, every time you ask for something, you get an answer back like it actually came from someone on your team. You set this up just once and from then on, anything that you would normally buy a tool for or hire someone for, you just ask. Now, I put together the full build that turns this one assistant into an entire team where every task is getting routed to the right specialist all automatically. [music] Comment "need" and I'll DM it straight to you.

여기 미친 남자는 뇌를 복지합니다. 평범한 메모를 AI가 읽을 수 있는 지식의 원천으로 받고 나보다 나를 더 잘 아는 퍼스널 OS를 구축한 지식 설계자 빈 그가 공개한 기록하는 습관 하나로 억대 가치의 아이디어를 뽑아내는 세 가지 필승 공식 저장하고 끝까지 보세요. 첫째 모든 기록을 마크다운 형식으로 남겨 AI 산소를 공급하세요. 사람들은 토큰이 중요하다고 하지만 사실 AI가 가장 맛있게 먹는 데이터는 마크다운 파일입니다. 당신의 모든 생각과 일상을이 형식으로 저장하면 AI는 당신의 기억을 완벽하게 복재 13개월 전에 사소한 메모조차 현재 강력한 무기로 되살려냅니다. 둘째, 나를 설명할 필요가 없는 컨텍스트 파일을 만드세요. 매번 AI에게 내 작업이나 취향을 설명하느라 시간을 버리는 건 아마추야입니다. 당신의 BS 철학, 프로젝트 목표, 심지어 선호하는 일처리 방식까지 담긴 전용 파일을 만들어 두면 AI는 이미 당신의 마음을 읽는 듯이 가장 정교한 정답을내어 놓습니다. 셋째, 그레드에이트 명령어로 죽은 메모에 생명력을 불어넣으세요. 1, 2장에 적어둔 낙서나 파편화된 아이디어들이 그냥 사라지게 두지 말고 AI에게 추출을 명령해야 합니다. AI가 당신의 전체 메모 창고를 뒤져 아홉 개의 핵심 태그와 숨겨진 통차를 찾아내고이를 단 몇 분 항장 실행 가능한 비즈니스 기획서로 자동으로 발전시켜 줍니다. 한시간짜리 영상을 모두 정리했습니다.이 이 자료 궁금한 분들은 구독 프로필 링크를 확인하세요.