Claude Code + Karparthy’s AutoResearch = GOD MODE

If you're using claude code without auto research, you might be leaving money on the table. So, in this video, I'm going to show you new strategy from Andre Kapi that could not only just improve any metric in your business, but also save you hours and get you light years ahead of your competitors. And if you don't know who I am, my name is Jack

s and get you light years ahead of your competitors. And if you don't know who I am, my name is Jack Roberts. I built and saw my last tech startup 60,000 customers and now I run an AI automation business. So, if you haven't already, grab that coffee and let's dive straight in. Broad code auto research can be best explained by WD40, the tool that

ee and let's dive straight in. Broad code auto research can be best explained by WD40, the tool that every man in the world needs. Now, what most people don't realize is it's called WD40 because WD39 failed. They had to do 40 different iterations of something before they got the actual winning combination. The TLDDR here is that the more we iterate

something before they got the actual winning combination. The TLDDR here is that the more we iterate on something and try and experiment, the better the thing can be, which can drive incredible benefits to our business or our personal lives. For example, if we work on the copy and the positioning, we can get higher email conversions. We can do the

mple, if we work on the copy and the positioning, we can get higher email conversions. We can do the exact same thing on our websites. Better title, better blogs, and get more sales. Likewise, if you want people to join your community or you want to increase your click-through rate on your social media or get better hooks or even sales scripts

t to increase your click-through rate on your social media or get better hooks or even sales scripts where your team are selling constantly or your clients are selling constantly, you name it. This process of iteration, gathering data, and making better decisions is what helps us to generally improve in anything we're doing in business. So what is

tter decisions is what helps us to generally improve in anything we're doing in business. So what is it and who the hell is this guy? Well, this is Andre Kapathy, one of the co-founders of Open AI. He also worked in I think it was with Tesla AI as well. Essentially, it is a GitHub repo that Andre has created that essentially enshrines some of these

ll. Essentially, it is a GitHub repo that Andre has created that essentially enshrines some of these core principles of this self-improvement cycle. It's blown up on GitHub. It's got many, many thousands and thousands of stars quite recently. But effectively, it has a couple of things. We just need something to change. We need an easy way to

But effectively, it has a couple of things. We just need something to change. We need an easy way to measure that thing. And then we just let our cla code agents run it and report back to us on a specified frequency that we give to it. And to do this we need three wonderful things. First of all we need one thing to change. We need one metric to

this we need three wonderful things. First of all we need one thing to change. We need one metric to measure and we need a way to read the result that it gives us. So the way this works essentially is a loop is we generate something. We deploy that thing and when that thing is deployed then we grab the data. We in effect harvest the data. The AI

t thing and when that thing is deployed then we grab the data. We in effect harvest the data. The AI are going to do this on a loop. It could be every 2 minutes. It could be every day. It could be whatever frequency we want to. But that's the triangular loop that we create something. We then see its performance and then based on that performance

ngular loop that we create something. We then see its performance and then based on that performance and data, we try a new experiment. The idea is that the actual quality of the thing goes up over and over time. And if you think about it this, day one, it's you're pushing, , a boulder up a mountain, a snowball mountain. But the core idea here is

one, it's you're pushing, , a boulder up a mountain, a snowball mountain. But the core idea here is that every iteration compounds and every experiment is logged and it never forgets. So we'll already know what we tried and didn't work. And so by day 365, you've got all this beautiful success that you could never have gotten before. Remember 1%

day 365, you've got all this beautiful success that you could never have gotten before. Remember 1% improvement. If you improve 100 by 1% every day, I think the total output is over 300. It's really impressive. So log is the asset essentially if you think about it this hand over from V1 to V2 model V1 compiles initial research and context that

you think about it this hand over from V1 to V2 model V1 compiles initial research and context that then passes over to V2. So V2 optimizes, enhances and builds on knowledge. Then V3 does the same thing and we keep giving this playbook of excellence together so we can figure out what the optimum is. Now if you think about this, the way that

ellence together so we can figure out what the optimum is. Now if you think about this, the way that essentially works is that we want some control, a control sample if possible. This won't always be possible and I'm going to show you a circumstance in which that might not be possible. But a typical idea is if we have two groups of things. We have

in which that might not be possible. But a typical idea is if we have two groups of things. We have a baseline control which is a blue card and then we have the challenger variant that claude code will create for us. And the idea with this is if the challenger variant beats the control it becomes the new baseline and so on and so on and so forth.

e challenger variant beats the control it becomes the new baseline and so on and so on and so forth. Add infinite item as they say. Now what makes a great candidate for this auto research process? Well, needs to be something that is objective. In other words, there can be no deliberation or debate about whether or not it was achieved. So, does it

ther words, there can be no deliberation or debate about whether or not it was achieved. So, does it feel good? Does it have a certain vibe? Not great. Is it a metric? Is it a conversion rate percentage? Is it something that's a one or a zero or a number or a pass or a fail? These are things that make it, but fundamentally needs to be an objective

umber or a pass or a fail? These are things that make it, but fundamentally needs to be an objective metric. And I will touch on that in the video as well, but also show you how you can come up with what you should be focusing on the very first project to do for your business or your personal life. Now, anything measurable can be optimized

rst project to do for your business or your personal life. Now, anything measurable can be optimized essentially. So it could be landing pages, email subject lines, ad creative, YouTube titles and thumbnails, e-commerce checkout flow, chatbot scripting. And the idea is that we build up all these different systems to run for us in parallel so we can

ng. And the idea is that we build up all these different systems to run for us in parallel so we can improve our entire business and workflows. And some very simple steps. We're going to clone the repo, write a program.md, define our metric, and then set the loop. And even if the thing that you want to track does not have an API, we're still going

n set the loop. And even if the thing that you want to track does not have an API, we're still going to cover that and show you exactly how to do it because APIs are not going to stop us here on the Jack Robbers YouTube channel. Wonderful. So now I'm in anti-gravity which is my IDE of choice, my integrated development environment. And then once you

in anti-gravity which is my IDE of choice, my integrated development environment. And then once you double click in the middle, you'll see an orange thing appear at the top. If you don't have claude installed, you can come on the left hand side to I believe it is just under extensions. You type in Claude, it's there. And then you download it and

I believe it is just under extensions. You type in Claude, it's there. And then you download it and you're going to be good to go. So we can choose how we want to connect Claude. I'm going to do it via my subscription because I've got a max plan. You can do it with the pro plan as well. Will save you on credit. So let's get that set up. We say

. You can do it with the pro plan as well. Will save you on credit. So let's get that set up. We say yes, Claude. That is absolutely fine. We're going to authorize you. And once you've done that, you'll see Claude appear on the hand side. And by this is by far the cheapest way to access Claude code is just get your pro plan. You don't need a max

his is by far the cheapest way to access Claude code is just get your pro plan. You don't need a max plan unless you're taking it into stellar which I think a lot of you are cuz you're watching this video and you're kicking butt. But ultimately the pro plan is going to be so much cheaper for you than the API. So this is the GitHub that we're going

pro plan is going to be so much cheaper for you than the API. So this is the GitHub that we're going to copy in there and it explains auto research how it works altogether and what it does. So the repo is deliberately kept small and only has three files that matter. prepare py fix constraints on one-time data prep trainy the single file that the

files that matter. prepare py fix constraints on one-time data prep trainy the single file that the agent edits and programm which is based on instructions for one agent point your agent here and let it go and it's got quick start instructions load of wonderful thing the best way to integrate this as with all these things is come over to the code

of wonderful thing the best way to integrate this as with all these things is come over to the code and we're just going to copy this and at the time recording it has over 44,000 stars so let's go back over then to anti-gravity and then let's give it here we're going to say hey there what I'd you to do is go ahead grab this repo and clone it into

it here we're going to say hey there what I'd you to do is go ahead grab this repo and clone it into this project folder please cool so it's just going to go straight over to GitHub, grab that, and drop it in. We're using GLD for speech to text. And if I come over here, you can see now in the explorer section, we're going to have everything's going

if I come over here, you can see now in the explorer section, we're going to have everything's going to be added. So, what we first of all want to do is just open up a folder. So, we're going to be rocking and rolling. You can also just clone a repo here, by the way, on the left hand side if you want to. You just put that down there, clone from

epo here, by the way, on the left hand side if you want to. You just put that down there, clone from this URL, and it will just automatically do it. And then we need to pick a folder, of course. So, if you come to any folder you want to, I can come down here, and I just say auto improve. Okay, cool. I hit create, and then select the repo's

an come down here, and I just say auto improve. Okay, cool. I hit create, and then select the repo's destination, and then it should clone the whole thing. Yeah, we're going to do that. We open that one up and then we have everything here rocking and rolling. There's an alternative option if you don't want to just yap to Claude. It's possible. As

nd rolling. There's an alternative option if you don't want to just yap to Claude. It's possible. As you can see, now we have everything there for us instantaneously on the left hand side. Now, this is the anti-gravity window which we can remove by doing control and L. You can speak with Claude if you want to within here. It's completely your call,

oing control and L. You can speak with Claude if you want to within here. It's completely your call, but we're going to be using Claude within the main section here by clicking in on this guy, which is wonderful. So, now we've done that, the first thing we need to do is essentially decide what is the thing that we want to track. Now, there's many

thing we need to do is essentially decide what is the thing that we want to track. Now, there's many things you could do. You could do things email conversion rates. We have different split things. Now, to apply these incredible auto research principles to your business, you only need to follow a very simple three-step framework that will

ch principles to your business, you only need to follow a very simple three-step framework that will essentially work for anything that you want to track and improve. The first one being choosing an important metric to change in the first place. And number one is something to change. So, let's take an example of the platform. Now, let's say for

And number one is something to change. So, let's take an example of the platform. Now, let's say for instance that you were on school. Now, many people are on school these days. So, let's say that you were on here and you wanted to track your conversions and you thought, , one of the things I want to do is see what my conversion rate is. So, let's

sions and you thought, , one of the things I want to do is see what my conversion rate is. So, let's say you have copy somewhere. Again, this could be on your website. The example doesn't matter. All that really matters is that you're picking a thing. So, you may look at this copy and say, "Hey, this copy is cool, but what if you could trial this

ng. So, you may look at this copy and say, "Hey, this copy is cool, but what if you could trial this on a weekly basis and have AI use this exact same auto research principle to try out different copies based on all of this information?" Well, that would be pretty cool. But in order to do that, we need a metric. We need a conversion metric. But the

ould be pretty cool. But in order to do that, we need a metric. We need a conversion metric. But the problem is not everything in the world connects via an API. If you're connecting to go high level or an email service, we can grab that. You can't grab that via school unless you use the strategy that I'm going to show you in this example, which is

rab that via school unless you use the strategy that I'm going to show you in this example, which is why remember the example doesn't matter. It's how we do it. So step one is decide the metric, which is going to be the thing we're changing, which is going to be this copy. Okay, it's going to copy all of this so. And now we're back into

ng, which is going to be this copy. Okay, it's going to copy all of this so. And now we're back into anti-gravity. First thing I wanted to do is to familiarize itself with the information on the left hand side. So, hey, the first thing I'd to do is to familiarize yourself with this entire project on the left hand side. And the spirit of what I'm

s to familiarize yourself with this entire project on the left hand side. And the spirit of what I'm trying to do, we are going to spin up several projects to track systematically improvements of a key metric in my businesses. And essentially, we're going to dictate what the thing is that we're tracking. I'm going to give you context about what

're going to dictate what the thing is that we're tracking. I'm going to give you context about what that thing is. You're going to ask me clarificatory questions. And you're going to follow along with the instructions on the left hand side. You're also going to go and create a claude.m file. And once we've done that, we're going to create a unique

lso going to go and create a claude.m file. And once we've done that, we're going to create a unique process for each of the things that we're going to track the actual thing we're changing the metric we want to get to and how what the process is of us improving that over time. So familiarize yourself with the items left hand side and let me know

f us improving that over time. So familiarize yourself with the items left hand side and let me know when you are ready to proceed. Awesome. So put this one in here so and then let Claude initialize itself with the information. Now this is wonderful. It's fully initialized and remember when you're deciding what you want to change, I will come on to

It's fully initialized and remember when you're deciding what you want to change, I will come on to that in this video. You're not going to want to miss that part of the video because it's really important is we typically want to pick the biggest constraining factor. What is the one number that exists in our world that if we were to change it

st constraining factor. What is the one number that exists in our world that if we were to change it would have the biggest impact in our business? I've chosen a random example but you'll get what I'm talking about. So now it's ready. We're going to decide in the first step what the thing is and give it crucially in this step context. Cuz without

o decide in the first step what the thing is and give it crucially in this step context. Cuz without context it's going to take way longer. Now obviously if you and I sat down optimize this every single day together we'd probably get there just as quick. But the idea is that we can automate more of it by bringing in claw code here. So, here's what

uick. But the idea is that we can automate more of it by bringing in claw code here. So, here's what I want you to do. I want you to optimize the copy on my school landing page. Your objective is to increase one number, which is the conversion rate number. The details that you're going to get fed into are going to be number of clicks and the