I spent the past three years consulting businesses on AI. And here's everything that you need to know about using it. First, just start with one job that your team already does every single day. So whether it's calling new leads back or chasing any unpaid invoices or typing the same job details into two systems that do not talk to each other, things that are occurring daily is the bar. Because a job that only runs 25 times a year, it's never going to be paying back what it costs to automate. And then from there, just write down four different numbers. the hours that it saves, what it costs to be connecting to the software that you already run, what it costs to keep it working every month, and what it costs to get your team to actually be using it. So, run this on one daily job this week, and you'll know in real dollars whether AI makes you money before you've even spent a single minute or scent on it. I've already built a sheet that does the math for you with real world price ranges for every number that you can't guess. Comment ne and I'll send it over to

There's an interesting pattern that I keep seeing with AI and it never starts with something going wrong. It actually starts with a win. Someone gets their first AI agent working and it feels like magic. So, they build 10 more and a few weeks later they come to me feeling completely overwhelmed. I run an AI consult and this comes up with almost every single one of my clients and the reason is almost always the same. Every agent that you build still needs someone to check it, fix it, and keep it on track. And that someone ends up being you. So, let me show you where that time actually goes and how to build agents that end up saving you time instead of taking it. Let's get into it. For those of you that are actively using AI on a day-to-day basis, the way that work is shaped is starting to change. In the past, you probably spent most of your time actually doing the work. So, you're doing the research, you're writing, you're building things. But now that you've adopted AI, AI is starting to do a lot of the work for you. So your job has shifted and now you're starting to turn into a manager, an orchestrator, or a reviewer. There's different words for this, but it's all the same thing. You have all these different AIs or agents that are actually doing work for you. What you're doing is you're reviewing their output to make sure that it met your expectations. And then this loop just continues. It does the work and you review the output. And what people don't realize is that as this shift occurs and you start to create many AIs that are working on your behalf, if you create too many incorrectly, the burden actually goes back to you and the benefit of AI starts to degrade. And that's where the overwhelm comes from. You have too many AIs working simultaneously. They're likely poorly created and that means you're having to do a lot of checking and analyzing afterwards. And that's the hidden cost. There actually five specific hidden costs that I want to highlight here for you. So the very first hidden cost actually happens before the agent ever even exists. And that's when you're explaining the work to the AI. If you explain the work to the AI poorly and your ask isn't very good, then you're likely going to have to continually keep explaining yourself over time as the AI doesn't meet your expectations over and over and over. That's the first one. So, this next one might seem a little weird, but I see it happen all the time where when you start an AI and you have it work for 10 minutes, 15 minutes, 30 minutes, even an hour, what you do is you end up actually watching the AI work, which counteracts the whole point of the AI doing the work for you in the first place. So, instead of you watching the AI work, you should go off and do something else and then come back when it's finished. So don't get distracted by watching the AI do its stuff. That's just another hidden cost. So the next one here is checking. So let's say that you built an AI and when you built it, it was built poorly. So when the output comes out, you have to keep checking the output because it's not meeting expectations. But that's the first part. The second part is that when you check the output, you have to read the entire thing. And again, that's going to take way too much time. So instead, we want the AI to create the output as well as a distilled version of it so we can quickly review that to make sure the output meets our expectations. And that's for some tasks, not all, but some. That's our third one. Our fourth one is cleaning things up. So this goes back to the point that AI can be overly verbose and also overly aggressive. So if you again don't constrain the AI in the right way, it could create too many files in too many different locations and you have to clean it up afterwards. Quick pause in your video. If you're enjoying this, you're going to enjoy two other things. First off, Blow is a 30-day AI insight series, completely free. You'll get 30 insights in your inbox of how you can apply AI to your business and your work. The second thing is if you'd like to work with me, Blow a series of offerings to see if there's a good fit between the two of us. Now, let's get back in the video. And then the final one is deciding. So, this is where the AI keeps asking for your input over and over and over, seeking decisions from you as well as answering questions for it. And depending on the task, we should actually outsource some of that judgment to the AI so we're not doing it ourselves. So, there's many different costs. These are the primary five that I've seen most frequently. And that's why people get overwhelmed when they have too many agents in the first place. But there's a way to fix this. There's specific rules that we need to follow when creating different agents. So when we're monitoring them and managing them over time, the burden isn't as high on us and we can get more leverage for these tools, not being overwhelmed. There are three primary c

So, Hermes's agents are great because it kind of feels like you have Claude code or Codex right in your pocket at all times. But, if you've ever tried to set one up before or you're intimidated to because you're worried about like the local setup or buying a VPS, then this video is for you because I'm going to show you how simple and easy this is with a managed Hermes's agent that you can get up and running by the end of this video. So, let's not waste any time and just get straight into this one. Okay, so real quick, I just want to cover what the difference is between a managed agent and a VPS Hermes's agent. So, if you don't care about this, then just go ahead and skip past this part of the video. So, at a glance, here are some of the quick hitters. The managed Hermes's starting off at about six bucks a month while the VPS can start off a little bit more than that. So, they're kind of the same price, but the managed is like slightly cheaper. There's also no server to buy or maintain on this side, whereas on the VPS you are buying and maintaining a server. It's still a relatively easy setup. It's still a one-click install, but there is more setup required. If you guys are interested in this version, then I have a video which I will tag right up here where I did it on a VPS. The setup is a lot more guided. It's a lot less intimidating, even though this still is pretty easy, but there definitely is more to configure and kind of like more to remember cuz on this side you don't have any Docker, no SSH, no ENV files, whereas this side you do, but it's still kind of builds you the container and the traffic. So, it's still not like you have to be super technical. You have full root access over here, full control over a lot of the config and anything else that you want on your server, whereas that's pretty much handled over here and you just have one managed agent in a box essentially. You've got Telegram pre-wired over here. You still get CLI access, and on the other side you have to set that up yourself. And you still own things like updates and backups and firewall rules and container health, but you don't have any extra cost really beyond the VPS, but you can also scale up the VPS later. Because really a VPS is not just for Hermes's. It's basically a private server, which you could scale up with more compute, more RAM. You could put other things in there. So, you can have a bunch of Hermes's agents in one VPS. You can have N8N in there. You can have a bunch of other things inside of this box, whereas with managed, you're basically just spinning up a new app every single time. You just get more flexibility with the VPS, but if you're not looking to be scaling up a server with tons of different things on it, then just start with a managed agent. It's easier and it's simple and it really will suffice for probably like 98% of you guys out there. But now that we're ready to actually get this set up, all you have to do is go to Hostinger. Go to hostinger.com, create an account, and then when you're in sort of your HPanel is what it's called, it will look like this. So, on the left-hand side you can see you can do things on Hostinger like websites, domains, emails, building sites. And in the past I've shown some methods of, you know, setting up a VPS and putting something on that like N8N and Hermes Agent and Open Claw and other things like that. But today I want to show you guys a method that's so much simpler and it's actually cheaper. All you have to do is come on the left-hand side and you'll see AI Agents. We can see Open Claw, Hermes, and N8N. So, obviously today we're going to click on Hermes Agent. You can see that I have two AI automation apps in here and these are both Hermes Agents. If I wanted to spin up Open Claw or N8N, I could do so right here as well, but we're going to click on Hermes Agent. We're going to click up top and go to get new app. And then all I have to do is click on get started for the Hermes Agent. And we're going to go ahead and choose the plan that we want for this Hermes Agent. If you want to include AI credits, you can, but I would just turn that off. I don't think you're going to use those. You could also get 1,000 free web scraping credits if you'd like and you can also get Agentic Mail. So, you can basically have an email address for your AI Agent. You could do this through Hostinger or you could also do this through something like Agent Mail. And by the way, if you use the link in the description, you can actually get taken to this landing page that will help you set up your Hermes Agent automation managed app and you can use coupon code Nate Herc to get 10% off if you're on a yearly plan. So, that has to be the 12-month or the 24-month, but you can use code Nate Herc here. It's the exact same product, it's just a different way to get there. So, make sure you use the link in the description and use the coupon code Nate Herc. And now this is going to take like a minute to get set