It almost feels like cheating. I feel that I'm getting dumber and laying easier. Like if I'm on a flight and there's no internet connection, I just don't feel like working anymore. That's Peter Yang. He spent a decade building products at Reddit, Meta, and Roblox. Now he runs a newsletter for 140,000 people with no team, almost all of it through AI. I'm like, like, an employee is never going to leave you. And it's only getting exponentially better. Last 10% you've got to add your human touch to it. And you don't have to be a creator to make this. I see a lot of creators, they get AI to generate, like, and post-prower, and like, the slot goes viral. This is heartening. For someone who's trying to stay ahead in this AI era, what should they be doing? So step number one is use codex or a clock code. And step two is to-- So I really wanted to make this podcast practical for everyone who's watching. There is something that's going on in AI right now that everyone is talking about as self-improving. And you recently build cloud self-improving skills. And the whole narrative is like, stop prompting your AI, make it figure out what to do next. Can you talk about that and is self-improving a reality for people who are just using AI versus researchers? Yeah, there's like different ways to do it. I think, first of all, I think all your audience knows the skills, just like a text file with a bunch of instructions. And yeah, I have skills for making my podcast for adding my newsletter posts and so on. Most basically, the way to do it self-improving is after you use the skill and you have a back and forth conversation with the AI, because it never gets a right in one shot. Then you just say, hey, based on our conversation, can you please update the skill to account for-- to try to get to it in one shot faster? And it'll make a bunch of changes. And then you should review it. And then hopefully, it'll do it better next time. When I integrated AI into every part of my work, it almost feels like cheating. And a year before that, you said you were drowning in meetings and projects. What are the top three things that you deployed in this year that dramatically changed how productive you are? So I can tell you my creator work. So we're both creators. And being a creator is a lot of repetitive work. A lot of repetitive copy and based and back and forth. And changing things from a newsletter post to a YouTube description to social posts is a lot of different formats. And I just decided to spend one day just like no meetings and just sat down with codecs and be like, hey, I basically just brain dumped all my workflows to codecs. And here's how I do a manually. And I use a thing called whisper, just like bringing up my voice. So there's like a couple major ones, like prepping the podcast, post production. There's editing my newsletter posts. There's posting to the various social media platforms. And then another one is like setting up a advisor to kind of give me like business advice, like checking ideas. - Is it like an agent that's running? - It's just like a skill that's linked to a Google dog file that has like a bunch of personal information about my business. And then it kind of gives me a vice. Talk to me about posting on different platforms. How does it work? Do you dictate something with your voice or does it see what kind of videos you produce and repurposes them? - I have some like a crown jobs, which is basically like, you know, like this game morning briefings weekly reports. That sent me how like other channels are doing, what kind of videos are popular. Now I post to social media platforms through codecs directly. Before I was using this tool called type 4 fully, these like social media tools. But like some of the platforms I post are like sub stack notes. They don't have any kind of like API integrations, right? So I just told codecs, hey, I want to schedule posts to sub stack notes can you help me figure it out. And then it went into the browser and it sniffed out some internal APIs that they're using to post. So it's a sniffed out internal APIs. And then it could also do computer use, right? - It could also do computers. Yeah, so you can do both. But now I can post to like four platforms. I can post to X, LinkedIn, threads and sub stack notes at the same time. And in the skill, it has like some nuances like for example on X, you want to tag the person that you interviewed, right? You want to find a handle. But on LinkedIn, you can already tag. It's just kind of hard to tag. So like I can just bring up the all the codecs and now I can actually do the tweaks for her for me. So it kind of sees me a lot of time. - Do you get the final approval before I post? Or it just goes ahead and post by itself? - I still draft it. I don't just get it to generate copy to post. So I still, I just like, you know, use what's before I bring, bring dumps and rat, rat, rat on thing into codecs. - So you talk to it, it takes your voice and just creates different social media posts. - Increasing social media posts. It also has my examples of posts that I've done that I've gone really viral. So it can help me like clear my brain dummy to like a more polished post. - Yeah. - And then I'll take a look and I'll be like, okay, go post it to all four platforms. - Does it have access to your analytics to assess what's going on right now? - It does, yes. It has access to analytics from various platforms and also, typefully, the thing I use also has analytics. So it is some analysis. Like for example, it found that on LinkedIn, I don't know if it's growing up, but like if you just do a post with a bunch of links to all your tutorials and stuff, it tends to do better than if you do a single one. So I had no idea about that. So I'm gonna do more of that now. - What about newsletter? Is it the same process where you talk about what you want to have a new newsletter and it creates it? - Newseller depends on what kind of post I'm making. So for example, like, I think by the time this goes out, I have published a post about why I'm leaving my job. So that one I just kind of like went on a walk and kind of dictated it to whisper to floor these voice tools and kind of just like bring down my real thoughts. - To codex, directly to codex. - No, no, I could do it to codex, but I use this tool called like Super Withsper that records like 10 minutes of my brain dump and then I copy and paste the whole thing into codex. 'Cause codex, I'm just worried like, if I take that for 10 minutes, codex will get confused. - That's my worry too. Or something stops working and it just keeps processing and nothing happens. That happened to me a couple of times. And it's not just codex. I think it's the context field that is not designed for a 10 minute prompt. And how many edits do you typically do in this process? - It's interesting 'cause like with AI, I'm kind of giving a feedback for it to edit. It's actually hard to just edit yourself and just give a feedback, right? So like I bring to help it, I give it a codex. It does a bunch of editing and it uses my best examples to edit. And then in order to do something wrong. So I have to re-thread and then I kind of voice takes a bunch of feedback and it doesn't our pass. And then maybe the last 10% I have to mowly change stuff. - Yeah. - It's actually important. I always emphasize like last 10% you gotta add your human touch to it. You can just like AI's lopify everything. - Basically all your written content is created with the help of AI. Talk to me about this strategic thing that you're doing. Is that like in the morning it briefs you on what's performing, what's not, or? - Every week it sends me a brief about how much money I made this week and also like how did my past 30 days content perform? And also like all the other channels that are mixed in their content, is there any kind of outliers? Does that for YouTube and also for Substack? And for Substack it doesn't have API so it has to use browser use to look up everything. - Yeah. - So that would take me an hour to do a man out of you but like it does it. - You mentioned some of your editing is done with AI. - Yeah. - Is that where you do dump transcript and it edits it out or is it a process? - Yeah, for video editing I feel like maybe I haven't gotten any interior yet but I still use like a human video editor. But the transcript is very useful 'cause I dump the transcript into codex and it tells me here's the moments where like you had like awkward pauses or like stuff you should cut. Here's the timestamps and also for like the intro here's kind of like we should pick like here's like the spicy quotes from the interview. - Nice. - Yeah, it's fascinating how this has become so much faster versus a year ago. - It's just yeah it's a different reality these days. - So you've described all of these things that you've built. How often do you build something new? - Pretty much every single time I thought AI is something changes, right? Like I think some skills or something changes but you're talking about like from zero to one. - Mm-hmm. - Well my hope is that I can streamline a lot of my content production and for the audience I'm not gonna slub everything. I'm still trying to taste but I wanna streamline a little bit so I can save more time because I actually wanna be a builder 'cause I spent a decade on my career just building products inside big companies and now that we have all these agents and tokens we can use, I wanna build different tools, open source tools or things that solve my own problems. Yeah, so I wanna do both. I don't wanna just be a creator, I wanna be a creator and a builder. - Nice. - So I'm gonna spend more time doing that. - How do you make decision when to build something and when it's okay the way it is? - It's usually, I feel like it's best to build stuff to solve your own problems. So for example, I built a fitness app. I mean everyone is building a fitness app but I built a fitness app that helps me trap my workouts and then it has an MCP to kind of send me emails on my health, that's one thing I built. I'm doing something else right now. You know in Gmail you get a lot of scammer's and my parents have no idea. Their bank sends them email but it's actually
[email protected], right? So I'm trying to build some sort of extension to just how high that this is a scam. - How long does it take to build something like that? - I think to get to 80% there it probably just takes a couple hours 'cause you just make a plan and you get caught up to it. But to actually make it good and to test it and polish it, that takes more time. That takes a couple days or like, you know, depending on where your quality bar is. - Yeah. - I'm quite aware of when I go. - Before we start building something, you have this system with five layers of AI adoption. Can you walk me through them? And let's decide which layer do we want to land on today. - I made this five layer thing actually like six months ago. So most people are in layer one which is just using AI for everyday answers. They're using chat GPT and cloud and asking questions. I think the second layer is like using AI for daily work. So you're still in chat GPT and cloud but you're using projects and other things to like, to not do too much copy and paste things. You like set up projects for different things like, I don't know, give me your life advice, all the many of your content stuff. But. But because it's in chat, you can then cloud, you still have to copy and paste the output to somewhere else. Right? So it's still like a lot of going back and forth. And then I think this is more from like a product perspective, but then a lot of people in my, you know, as a product manager, you want to use AI for prototype typing. Because like product managers, we all use just a lot of dog arguments that not very engineers want to actually read. So now we've got all these tools like level bow and like, um, replete and, you know, codex and everything else, you can actually build prototypes of your products. So like I usually take a screenshot of my product and I'm like, Hey, can you make a prototype of this and make some changes and show it to my engineer and the designer? Uh, so that's then, and then you can use AI level four to build different apps. So like, there's been been building a personal app for yourself and trying to like build an app that like millions of people use, right? So when I say building apps, it's a lot of it's just like personal apps for yourself. Does eight, where do the agents go in these layers? Well, if you just build an app, like, you know, you can just print on a verse or something and then have a link that people can use. So it's not really looping or anything. Okay. So that's layer four and layer five layer five layer five is what we've been talking about, which is like you actually use AI kind of for like as a personal agent to try to like stream on an automated as much as you can, right? And like, and it sounds like very complicated, but really it's just about like sitting down, like, like, canceling our meetings for a day and staying down and like thinking through your past week, all the time you spend time manually doing stuff and just like sharing it with your AI friend. Can you walk me through a process that we just talked about in a very practical way so that people who've been using AI on like level one, maybe some level two, can move to maybe ideally level five, where it's more. I feel like the most practical step that most people should take is just to like stop using chat, chb10, and just start using codex or a claw code. That's like step, step one. And like people get intimidated by this stuff like code isn't claw code like this is like a programming. Sounds very nerdy. Yeah. But in reality, it's just like you just do is just chatting with the AI, right? And like when I use codex like 80% of time, I'm not building, I'm not coding or doing that. It's just more powerful because, you know, these two apps, they have, you can build skills for them. They have access to APIs and integration so they can actually get work done across your other apps. And they're just like way more powerful than just default chat. You can't cloud. So basically like what I recommend you do is use codex or a claw code and set up a new folder or call it like personal OS or whatever you want to call it. Just call it a folder, right? And then I asked it like what are the main workflows that it can do here? And you can kind of see here that, um, oh, you asked codex to come up with. Well, this is all the stuff that I've set up over time, right? So basically like, you know, as we talked about, it uses like various different skills and skills are just text files to do work for me. So like for example, to edit my newsletter, it has this skill to shape a rough note into transcript. And then it also uses other skill called last three days to add fresh research and then it uses another skill called no AI Slop to make the writing clear. I like that. What if you don't ask it for the last 30 days? It's just going to admit. I think it's always in the skill. I have so many instructions, but like, hey, you do some all-in search. Just to make sure I'm like, you know, saying stuff that says true. Another thing people should recognize is that there's a lot of like open-source skills that are really good. You know, so last 30 days is like some open-source skill that I found on GitHub. And it does like searches across like reddit, x and all these other platforms. And then it uses no AI Slop to and then eventually the newsletter post gets to a point. And I use social writer skill to make to turn newsletter into like a bunch of threats. Right? So that's for newsletter. Your podcast is very similar format. Like I use this podcast prep skill to research guests that I have. And then, you know, as part of making stuff on YouTube, you know, it's better than anyone. You're going to have a really good thumbnail title and call it P. So I have a whole skill to just like test different thumbnails and titles. And then based on that, I decided the podcast is even worth doing. And then I do like a post production skill, like turning into all kinds of stuff. Yeah, sponsors. We both have sponsors. It's like a lot of work. So I asked it to keep track of my sponsors through different skills and also help me draft and brainstorm sponsor copy, help me, you know, schedule stuff on the inventory and all that kinds of stuff. Yeah. So so I guess the TRDR is like these workflows are done by connecting different skills together that you build. And like for all these skills, I don't actually. Can you show one of them like just go into like strategy, for example, where it's built and like walk through how you built it? Yeah, sure. I think one of the most useful skills that I built is just like a personal advisor skill because I guess my wife is tired of me asking for advice all the time. It literally is just a text file, right? So it's a text file. It's giving me an honest advice and you know, you're my trusted life and business advice. And then keep your tone warm, right? And then this stuff is like a little bit more fancy. There's like e-vows and stuff, but we can talk about that later. Learning is that your file? Yeah, as Codex itself has memory, but I asked it to save an error text file called learnings.md to save memory from our past conversation sessions. So it kind of learns about me. This is kind of the self-improving stuff that we talked about. With something they've released yesterday, because they released something when it now remembers all of your conversations like has this expanded memory. Do people still need learnings? Indeed. Well, I don't actually know how their like system wide memory works. So maybe don't need it. But like it's also useful for me to read learnings.md and see what kind of like, uh, conversations I've had with it in in a past. For learnings, md file, do you ask it? How do you prompt so that every conversation contributes to that file? Yeah. So, um, you also have to think about managing the context window because you don't want to be like super long. Yeah. Right? So I have some instructions saying like, Hey, like after every conversation you should ask me, like, you should recognize kind of like patterns and like learnings that you have and ask me if, if I want to save and learnings that md. So ask me. And there are more strategic learnings, right? Because this is your personal advisor. So after every single conversation, uh, issue uses is best judgment. What it wants to ask me or not. And then I also ask it like if you're going to save it there, just like have a few lines. Just like have one or two sentences. Like don't don't like huge paragraphs. So that's the way it's so interesting with these like how what what AI actually remembers about you. Yeah. Because sometimes you think maybe remembers everything and you hope that it does, but in reality, it doesn't or remember stuff that you don't really care about. Yeah. So you basically prompted it saying, what where does this prompt go? Does it go? It's just in the skill file, like the main skill file of like, Hey, before you before you answer any questions I have, like just quickly skim through learnings.md. And once we've finished conversation or like in the middle of conversation, we see something ask me if this should be added to this learnings and defile and the skill runs across everything that you do. The skill is triggered based on like your description, your skill description in the top, which is like, so here I say like use whenever the user is stuck on a decision or working through a hard problem or asking for a gut check. But sometimes it doesn't trigger on sometimes that you like actually do like slash, give me a slash personal advisor. Yeah. And then ask a question. How do you come up with all the like for me when I'm building this? Yeah. There is so much brain work going into this like when a trigger the skill, all the nuances. Do you use codecs for that as well? Yeah, I never write any of these files man, man, only. But it's still, no, it's even not writing. It's just thinking about all the nuances. Yeah, yeah. Yes, I can think about a lot of things for you. But then there are so many different things that you want to remember like when to trigger the skill. What types of information you want to go there? Oh, maybe. And then you wake up in the middle of the night. Oh, I forgot to add this. That's right. So yeah, I have a skill to build skills, right? And I thought it actually has a skill to build skills. Yeah. Personally, I'm like very worried about just like it becoming a sloppy fight. Like it just like starts adding stuff. I don't ever read it. And it becomes like super sloppy. Like it becomes like super sloppy. Yeah, because you also want to make sure you're in control of whatever is in the skill. And if AI just built a huge skill for you, then you don't have time to read it. So my, so my, I have like a skill editor skill is very basic. It just like looks for like duplicate instructions or like a slop content and try to remove it. Like, like I tell you to keep my all my skills to like one page max. So like as you read through it, it's even a sense of that. Let's take a deeper look at what's next. Yeah. So, so yeah, so like update the plan and then, and then like so for this personal advisor thing, I have my, you know what, why don't I just ask it to you? Can you read my plan, Google Doc and share its overall structure without sharing confidential info? So basically like, and this is just the way I do it. Like I have my actual plan or like, you know, my business plan or whatever in the, you know, Google Doc and like I find it's very useful to as creator to like have some principles to help you make decisions. So basically it's like a one-pager with my goal, my principles, like background my business, positioning and some other stuff like energy and other stuff, right? Like I don't know why you're a goal, but Michael is just to like make X amount of money and just like actually being control my time and who I want to be with. And then my principles are like, like I have a problem where I tend to take on too much stuff. So I asked you to like just keep the main thing the main thing, like that's one of principles, like which is like, you know, people invite you to conferences and like, you know, do all kind of like try to write, write a book or do all kinds of stuff and just, and you have to say no because you want to, you should focus on making a newsletter and a podcast great. Yeah. It's so hard to. Yeah. You focus, right? Yeah. But because I have that principle, so when it gives me a vice, it has that principle in mind. And then the business stuff is more just like, you know, like all the financial information. And I think it's a very, why important section is like what gives me energy and what takes energy away and just like a couple of bullet points. But like, you know, what gives me energy is like, you know, vibe coding with codex.
What takes energy away is like, I don't know, like doing too many Zoom calls or something. - Course totally. - Yeah, so like, so I think it's very important. Like you don't have to be a creator to make this document, right? For any kind of career you have, just make this document and putting a Google Doc somewhere and then set up a skill to reference this document when it gives you a vice. And I think that like it's literally one of my most useful skills is like not super complicated, but like it helps and makes love decisions for better for worse. Most people use Claude like a search engine. They type in a question, they get an answer. Most times they're not really satisfied with it and they close the tap. I did the same thing for months and I was looking at people who were saying AI is changing their life and I'm like, "Hmm?" Then I spend one afternoon setting it up properly, uploaded a few files about how I think and how I work and it completely changed. I wrote the whole process up step by step. You get it when you subscribe to my newsletter, Future Proof. It's free, the link is in the description. Yeah, this is very powerful because then you can use it for anything that you do so it adheres to your principles. Do you have a team? I have a video editor and a few contractors, but no, I don't have a full time team. Have you ever tried building like an AI chief of staff? Yeah, I have one in this thing. We're not using it this in called Irmiss. Yeah. Can you talk about this? It's like open-call, but it's like a little bit more reliable. And there's like a team that built it. But you can also just set up in codex too. But I'm using Hermes right now because I like having in my messaging apps like in a program and like wherever I want to use it. How do you use it? Talk to me about that. So I'm trying to build something. I'm trying to build an AI chief of staff where, first of all, my strategic decision making happens on Zoom calls and in Telegram where I give feedback to my team and I say yes or no. So I want AI to learn my strategic decision making style. Got it. And then I just talked to you know, I interviewed the front of Kalshi. And she has a team of developers, but they build this thing where it goes across all the slack chats they have every Sunday. And tells her where the team is stuck, what was promised to be delivered, what hasn't been delivered. So somebody who just watches the whole operation and gets everyone accountable. I see. Were you trying to build it with Irmiss? So you just have to think about where the inputs, what are the, what are you trying to do and what are the outputs, right? So the inputs for like any kind of business is probably like a slack. Yeah. And if you use your knowledge to require your meetings, there's transcripts there. So you just set up all the, just ask like codex or Irmiss or whatever. Just set up all the integrations. But like, hey, can you hook up to my granola? Can you hook up to my slack? So for anyone who's at level two, let's say, and wants to get to level five, what are the steps they should be taking? I think this is the set number one. Really build something like a strategic mindset. Behind all of your AIs and projects and agents, what are the next steps? So step number one is, as I said, you just actually use codex or a cloud code. Just start using that instead of like the regular stuff, right? Even projects like you do. Yeah, I switched away from projects because like I set up all my projects here instead because like, like with a project, if I want to edit and use that or post, it's spit out an output and I have to copy and paste the output to some type. It's a Google doc, yeah. Or like, mainly update stuff. But this stuff can just update Google doc for me. So it saves the step, right? Makes sense. So no, if someone is a user, right? Apps and step two is to just ask the right questions. Like bring down your workflows and ask the right questions because I feel like a lot of people don't realize what this stuff is capable of doing. You know, like this stuff is basically capable of doing in my opinion, any kind of knowledge work that you have, as long as you're patient with it and give it the right context. So just ask the right questions and then set up the workflows, like we just talked about, and then build the skills and the integrations and then you're off to the races. - What do you think your life is going to look like in a year? - A year. I've always worked in big companies where there's always being a lot of cross-functional alignment, right? You got a lot of different teams and different orgs. I really actually do really miss my engineering team and people like that I enjoy working with at work. But I feel like these days, like this is like the era of solar burners and I feel like you can get so much time with these agents, stuff that you don't want to do. So it really frees up your time to do stuff that gives you more energy. And even the stuff that's strategic would give me energy, I feel like now I have this partner that knows me and can be my pair programmer or partner to kind of figure stuff out. - Do you think you're gonna be working less or more in a year? - Well, I think I'm pretty type A, so I don't really think I'm gonna completely check out. But I hope I'll be doing more work that I enjoy. I don't think I feel like if you and I retire, it would be a bore over my employee, go play golf or something, but it would be very bore. - Of course, yeah, I think we're gonna keep working. - Yeah, we're keep working, but we want to do work that we actually enjoy, not like the Montenegro's copy and paste or like, you know, that kind of work. - Do you have any fear around AI? - I feel like I fear that I'm getting dumber and a lazy, like if I'm on a flight and there's no like internet connection, like that, I just don't feel like working anymore. I don't have a partner. - You know that's evolution. If you take our grandparents, even my grandparents, like they would never understand my lifestyle 'cause they were constantly working. It's either home like my mom raised me while working, while cooking, while cleaning, and I'm like, oh, I can't clean, I have, I need a cleaner. - You need a cleaner. - I'm like, oh, I just bought a robot vacuum cleaner and I don't know how I lived without it. We're becoming laser and laser, that's true. But I think it's, it's actually, then we have more time for kids and meaningful time with the loved ones. - I do worry about our kids 'cause like, I grew up without AI, so we have like 20 years of doing the basic stuff, like basic critical thinking skills and like, you know, learning how to do math and stuff. But I feel like our kids these days, they can start using codecs right now. They can talk to a computer. So like, you kind of have to make sure they actually still learn their fundamentals. - What's your solution to that? What are you trying to do with your kids that's different from how you were? - I have kind of like a radical thing. I feel like they need to, and maybe this is my bias, but I feel like they need to just like, maybe start business or something. I'm learning the real things. 'Cause I feel like, reflecting on my journey, my education, I didn't learn time from the classroom. I learn by like failing and like making mistakes. - Yeah. - So I feel like they need to have the right mentality to do that kind of stuff and like learn how to fail and just go out there and like try to do something. - Do you ever have fear for your career? - Yeah, my career? - Yeah, I'm talking to a lot of creators these days and I see them at events and a lot of people think that creator economy is getting over-saturated 'cause everyone is suddenly creator 'cause it's so much easier to create content. Do you think that might be a problem? - I do think it's a problem 'cause I see a lot of creators like just making a slap, like they get AI to generate like 10 posts per hour and like the slap goes viral and it's kind of like this is heartening. Do you see that? But I feel like that, I think that's why it's so important to find your principles and values. 'Cause like I never wanna be someone who just makes a tons of slap even if it goes viral it makes me a lot of money. - Yeah. - But I do think, but maybe people will get tired of a slap and people will be like, hey, I actually want human craft and human touch on this stuff. - Let's finish with an action plan. - Yeah. - For someone who's trying to stay ahead in this AI era for the next week, what should they be doing? - Okay, I think to stay ahead, get the $20 chat GPT plan, download codex and just start saving time and building workflows and building skills to save time and asking players to do it. 'Cause I feel like that's more productive for you 'cause you immediately recognize that you can start saving time versus what a lot of people do which is like they go on Twitter and try to read all the news and try to take some courses or something. Like switch from consumption mode to actual build mode. - What if they come back and say, oh, I tried, but it didn't work. I don't know where you're talking about it so bad. It's so much easier to just do everything for yourself. - Well, it does take longer, so it's interesting 'cause like with AI, you have to settle the system first before you can see the fruits of the output and it does take time to settle the system. But it probably only takes like one day to do it, right? And you have to have some patience 'cause even now with all this stuff I built, it still cannot one shot exactly what I want. So there's always kind of a back and forth and as long as you're patient and you give it like feedback and tell how to save it in your skills and memory, it'll get better over time. And unlike an employee who's never gonna leave you, it's gonna be there. - Never gonna leave you and it's only getting exponentially better. - Yeah, the mall will get better. - Single month. Peter, thank you so much. It was a very, very productive conversation. - Thanks for being on. - Thank you for coming.
PodcastTranscript AI