Frey Chu

where it gets a little bit more interesting.

And this was like where I had the biggest problem a couple years back, which is, okay, we have 20,000 potential port-a-potty businesses

I've kind of exhausted all of my ways to kind of superficially clean the data.

And so at this point, I need a way to automate the process of manually going to every single website and verifying whether or not these are luxury restroom trailers.

And so the way that I did that is I installed Crawl for AI.

This is an open source, LLM friendly web crawler and scraper, totally free, which is insane.

And

This is something that you just install locally to your computer.

I'm brand new to AI coding.

I'm only six months in.

And so if you're just getting started, the way that I literally did it is just took this link, gave it to Cloud Code, and told it to help me install it.

And it took about 15 minutes.

Now that we have this installed, Crawl for AI is kind of the engine that allows us to, at scale, look at every single website.

And then Cloud Code is the brain.

So we can prompt Cloud Code and say, look at every single one of these 20,000 websites and just go and identify the luxury restroom trailers.

So as a quick little demo here, I'll just pull up my...

Cloud Code here.

And I just wanted to show people the workflow instead of just talking about it.

But I'll just do it on a smaller sort of sample size.

So here's the CSV of just 10 restroom trailers, or sorry, here's an example of 10 porta potty companies.