Build AI Agents with Three Fundamentals
A practical framework for designing agent systems around memory, model inference, and the capabilities that make action possible.
Where to start
Start with the free Reality Check. If something’s worth building, the Guided Build comes next.
-
The Reality Check
Free
Forty-five minutes on one process that takes up your week. You leave knowing whether it's worth automating, what it would take, and whether I'm someone you'd want to work with.
If AI isn't the answer for it, I'll tell you so.
-
The Guided Build
Fixed price
The process from your Reality Check, built into a working AI agent by your own people, with me, in six weeks.
It runs on your real work in six weeks, or you don’t pay.
Rather have it built for you, or want help on an ongoing basis? Tell me what you have in mind.
Bio
I started this work to serve. I believe the complex can be simplified, and that understanding comes to those who seek it. I come alongside the people who do the work and help them use what's new in AI where it actually helps.
I'm Dillon McClendon. I studied mechanical engineering at the University of Tennessee, Knoxville, and spent the first six and a half years of my career at Denso Manufacturing Tennessee, as a production engineer, then a manufacturing systems engineer, then a software developer. I put in computer vision systems, wrote software for custom production machines, programmed robots, brought in autonomous robots to move material, and built manufacturing execution software. Building software applications has been my full-time work since 2021.
I've used AI chat tools in my own work since they first came out in late 2022, to learn faster and write code, and for the past year I've run AI agents as a core part of how I build software. At work, I built an internal tool to speed up my own software development, not a product: it ran AI agents through long tasks, gave each one a focused job, checked the work at set points, and rethought the plan when things drifted. It did real work for a short time, until the big commercial AI models got cheap enough to do it better, and I switched.
On my own, I'm building a tool that shows what an AI agent actually did: what it read, what it missed, what it cost, and whether anything left your computer. It's early, and it runs on small models on my own machine.
Outside of work, I'm with my family, my church and my friends, on a DIY project with my wife, learning something new, or, once in a while, snowboarding.
Free reading
How these systems actually work
Free notes and guides on how AI works. Start with the basics, and use what you learn with the tools you already have.
-
AI and Agents in Plain Terms
Working definitions of artificial intelligence and AI agents, the corrections those definitions usually need, and what an agent can and cannot do.
-
What a Model Actually Is
A plain-language account of what an AI model is made of, where it came from, what it does when you use it, and the vocabulary that comes with it.
Start a conversation
Email me the process you have in mind, and I'll tell you whether it's worth a closer look.