Some might assume that my 30-plus years as a small business computer consultant would give me an advantage in learning agentic AI. It did not. My first encounter with the OpenClaw agentic system was as challenging and frustrating as if I had never installed an operating system or configured an accounting or database system. It is an entirely different experience. My point is that you do not need to be a "computer aficionado" to benefit from the enormous advantages of becoming proficient in the use of agents (or bots, as they are sometimes affectionately known).
I was born at the tail end of the 1940s. I will let you do the math. My point is, I am hardly a young computer nerd. Like nearly everyone, I was introduced to AI via ChatGPT, Gemini, Copilot — take your pick. It was likely some version of a "chatbot." Agentic AI is not that. What I found was that, although chatbots were like search tools on steroids, they lacked the ability to handle long, multi-part projects over time. Their ability to remember and connect was severely limited. That is when I heard about agents. Agents could remember, handle multi-part projects spanning long periods of time, and could interact with the real world through the use of tools, skills, hooks, and other means.
But I quickly found that not all bots are created equal. After months of attempting to do real work with OpenClaw, I began to recognize some serious design differences that conflicted with my goals — in other words, I couldn't get it to do what I wanted. That was why I became open to finding a better or different agentic system, which brings us full circle to Hermes.
Both OpenClaw and Hermes are open source, so licensing isn't the difference. Where they really part ways is philosophy and design. OpenClaw is built to run locally, on your own hardware, with a broad ecosystem of community-built skills — a powerful tool you configure to your liking. Hermes, by contrast, is designed around a learning loop: the more you work with it, the more it learns about you, your style, your preferences, and the projects you prefer to work on. In other words, where working with OpenClaw felt more like a master-and-servant arrangement, working with a Hermes bot is more like having a willing partner — or at least a capable assistant.
The purpose of this article was to introduce you to my personal journey into the world of agentic AI. It is my hope that each of you will find your own, individual way of reaping the benefits of a system like Hermes. You may find that, for your particular purposes, OpenClaw is the better choice. I encourage you to investigate all the options. In the end, this site and the Windmill Club are intended to be an introduction to the world of agentic AI. I hope that you will find your journey to be as rewarding as mine has been.
