The Wisdom Wall

“And a new way to do it is actually have the LLMs just write the code you need right now. And then you write 10 or 20 times less code, but it's way more, uh, customizable. It does, you know, more and is more awesome. Between token maxing and even like, you know, don't write code that calls LLMs, write markdown that…”

“And then now it's like, actually what would take a week or a month or a year, like you literally do in, you know, minutes, hours, or, you know, at the worst, like you could build a whole cathedral of like 20 to 40,000 lines of code. Like what would take a typical engineer an entire year? You do it in a few days, right?…”

“how do you build super intelligence inside a company? You do that on everything you do. And it's not more complicated than that. Like, you literally just compose everything that you do, and any given thing that any given person can do, you combine that in aggregate, and in this particular process, and, like, you have a…”

“All of the difficulty in energetic engineering today is when people try to do things that should be in markdown in code and it fails because code is brittle.”

“It's very significant for people who are technical because it actually raises the bar on like what you are capable of doing. Like all the people who are attacking me about lines of code, they particularly are, The people who are most likely to get wings if you, like, let it rip and token max. This is sort of like the…”

“One of the key maxims for YC is You know, how do you find good startup ideas, live in the future and build what's missing? Right. And so this is a profound version of that where all you have to do is commit your brain to look at, you know, spending 500 dollars in a single day on tokens and say, actually, like, you…”

“The bottleneck here is not Not the model's intelligence. As long as you set the models up right, they are already smart enough to do extraordinary work on your code base.”

“When you're paying someone else, like the reason why that's interesting for consumer products is, uh, it's a mispriced asset generally still. I mean, Mr. Beast is not a mispriced asset. He's like getting his value from it. Like he's, you know, but It does seem like the mispriced assets are sort of the creators with…”

“Um, you, you can almost Argue that anything that is consumer and publicly available on the internet, that's going to be in the base model. So then your moat ultimately is for all of the other things that are not already online, whether it's, you know, for case techs being a lawyer, or maybe over here on science, or,…”

“If you treat OpenAI as the Google of the next 20 years, you want to invest in Google and all the things that Google enabled, like Airbnb. Google could do Airbnb, it probably won't. Just from, like, I don't know, Coase's theorem of the firm, probably. It's just, like, too inefficient and too difficult, requires too…”

“while in the next five or 10 years I think we will get far more used to using it that way, um, I think the, the low-hanging fruit right now is just using the large language model To actually do the sort of knowledge work that a human being could do, and then package it into the UI that, you know, whether it's a mobile…”

“a bunch of us are funding, uh, dev tools companies that sell to AI companies and they're selling tooling, but then they might, you know, they might sell, uh, enterprise contract to someone who also upstream has a fortune 100 that said that they'd pay a 100,000 dollars a year for that contract. And then six to nine…”

“Anything you do with those prompts, you can get your own model to do with a little bit more training.”

“I mean, this is why I think the chat interface is wrong. Like I actually think there is value accrued to really great UX, like good copy, good, um, you know, interaction design, information hierarchy, uh, you know, being able to approach a product and say like, this is the job to be done. And for users to come in and…”

“these things are really good at working with MCP and CLI. Like they're a little even better with CLI.”

“that's why you should start a startup, because people are going to be trapped in organizations with people running organizations that are very powerful and have all these resources and all this capital that do not believe what we just said.”

“unless you have your own prompts and you can write it for yourself, like you are You know, below the API line for some PM or developer that is not you who like will not understand you will not understand your needs will not understand what you uniquely care about.”

“now people are probably just making Architecture decisions about what to use directly in CloudCode. Like, they might not even know what, you know, what analytics to use. And it's like, oh yeah, as long as CloudCode says use PostHog, like, they're using PostHog.”

“I think memory is turning into, um, an actual moat for like that consumer experience.”

“Well, it turns out that 90% of enterprises don't know how to do You know, IT, let alone AI. It's weird to say that that's a good thing, but in the context of fast takeoff, like that is a real break on the ability of this new, really insane technology from actually permeating society.”

“you can't get the evals unless you are sitting literally side by side with people who are doing X, Y, or Z knowledge work. You know, you need to sit next to the tractor sales regional manager and understand, well, you know, this person cares about, you know, this is how they get promoted. This is what they care about.…”

“back end ends up being about actually infrastructure, and then front end is so much more actually being a PM. You're sort of, um, almost being like an ethnographer going into the obscure, underserved parts of the pie of GDP, and you're trying to extract out, like, this is what those people in that GDP pie actually…”

“once you're in the land of, um, doing image gen and video gen, there's just a lot you have to do around curation and proper prompting. It's kind of like a dark art, actually, so being able to start off with the thousand examples that you've already pulled together, I mean, that's context in action.”

“Interestingly, like we're sort of entering this other moment where, because ploy exists, you know, you might have someone who's like, 200 IQ, uh, you know, nearly nonverbal, like, codexes. They're able to make some sort of hardware, software that literally no one else could. But in the past, like, if you're that You…”

“My answer from, like, the OpenClaw Hermes experience with G-Brain is, like, yeah, you basically have to take that you're gonna denormalize it, and you're gonna put it in a format that, Is optimized for agent retrieval and understanding.”

“And so if you have a dry and MECE resolver table anywhere, it's actually like the optimal resolver Like, it's bad to have 10 skills that do all the same thing. It's good to have one skill or one tool that has parameters that then let you call them.”

“And they, like, sold 500 of these Apple Ones, and I think we're at the Apple One moment right now. We are coming up with the primitives, we're learning how do these things work, and how do we sell it, and how do we package it”

“One of the things I discovered is, um, sometimes Claude would just get confused and like write bugs or not be complete. But once I started saying, actually, before you start your work, make an ASCII diagram of all the data flows, all the inputs and outputs. What are the user flows? What are the error messages? And you…”

“I've since learned that a hundred percent is probably too much. Like hitting 80 to 90% is usually the best practice at this point.”

“One is, uh, what is the 10 X check? What is more ambitious and delivers 10 X more value, uh, for only two X the effort. Right. And so for whatever reason, coming out of latent space, this helps the model, like really visualize”

“Like if it's not tested and you're just throwing users in there, like it's slop. You know, 10 X worse than like human written code.”

“Uh, and then if you can token max, it's like, I mean, you can buy millions of years of consciousness of machine consciousness. Now I can be a time billionaire. It's not, you know, my own time. It's the time of a machine, like doing work for me and like the human entities that I care about.”

“It turns out, the way to get agents to do real work is the same way humans have always done it. As a team, with roles, with process, with review.”

“Out of the box, the model wanders. It doesn't know your data well, so it guesses. And guessing at that scale is how you get plausible-looking code that silently breaks.”

“the explosion in consumer in, you know, sort of the 2008 to 2012, 2014 period, that was an opening up of platforms. And then as the consolidation happened, you know, distribution closed down, which sort of spurred, you know, this move over to B to B.”

“a much more powerful version of this is let's work backwards from what I want to put on X. I'm going to make a very simple Loom video showing off a feat of strength, a thing that I really want to share that I know our team can do, and then working backwards from that, the next two weeks, that's all we're going to do.…”

“Like objectively, you should make literally a spreadsheet. You should, you know, go down and like evaluate it the way an investor would. And then the difference is the investor has a portfolio and you just have one life. And likewise, when you start a startup, you should not try to start a startup just to be the median…”

“I mean, I think you want, I don't know, at least six, maybe nine months minimum of like, just, you know, being able to live on ramen in the cheapest possible way.”

“Like, you definitely can't farm this out. Like, literally, the founders themselves, they're technical. They have to be the great product people. They have to be the ethnographer. They have to be the designer. You want the person on the second meeting to see the demo you put together based on the stuff you heard, and…”

“It also kind of feels like learning how to manage a person. Where it's like, how do I actually communicate, uh, you know, the things that they need to know in order to make a good decision? And how do I make sure that they know, um, you know, how I'm going to evaluate and score them?”

“The people who are the absolute best at improving the process are the people actually doing it, and it's literally why, ah, Japanese cars got so good in the nineties, and that's meta-prompting to me.”

“My point of view is you get more breadth now in terms of the things that you can try, but it still requires a lot of depth to fully explore it.”

“you shouldn't use like VC thinking to decide what to work on. Cause a lot of people Might have looked at how many designers are there in the country, and then multiplied that out by, like, whatever they thought they could spend. And it's like, oh, that, you know, that number is too small. I shouldn't work on that.”

“A lot of times, we as group partners aren't totally sure about an idea at the interview stage. It's really about identifying these key traits, the same ones we've seen time and again in the most successful founders.”

“When you don't launch or when you wait too long to launch, you are missing out and delaying the actual learning that you could be having by having real people reacting to what you're doing.”

“Companies like Apple and Google have infinite budgets. They spend years developing new products and millions of dollars on each launch. If you applied that to any of your startup, your company would die long before you launched anything.”

“Those 100 people who love your product become your viral sales force.”

“If someone is willing to pay for the jankiest version of your product, you'll know You're solving a real problem.”

“first you have to let someone know what the heck it is. And then right after that, you have to make sure that they know that what you're doing is awesome in some way, like that you're worth spending, like, you know, that was 10:15 seconds, and then it's worth at least a minute conversation. And if you can make it to a…”

“the model really wants to actually help you so much that if you just tell it, give me back output in this particular format, even if it doesn't quite have the information it needs, it'll actually just tell you what it thinks you want to hear. And it's literally a hallucination. So one thing they discovered is that you…”