About

I'm Patrick Sawyer, and I've spent about twenty years in consulting, doing everything from business analysis and data integration to delivery leadership and product ownership over software development teams, for clients ranging from Fortune 25 retailers and DoD organizations to small and medium businesses.

Before any of that I worked retail as a store associate, and I liked it more than people expect. Stuff comes in the back, goes out the front, and there's something satisfying about a job that simple. I got serious about supply chain in school right around the time one major retailer figured out how to make a barcode scan at checkout trigger a production order at a factory on the other side of the world, in near real time. Nobody else could do that yet. That's the first time I really understood how one shift in technology can rewrite an entire industry's rules in a few years.

I think AI is the next version of that same kind of shift, except bigger and faster. The supply chain revolution took a decade or more to reach most businesses because the infrastructure had to catch up: compute, bandwidth, systems that could talk to each other. AI doesn't have that problem. The cloud, the compute, the connections, all of it's already built. There's nothing standing between "this could work" and "this is running" except someone who knows how to build it. I think that makes this the fastest, biggest shift I'll see in my career, and I'm genuinely excited to be building through it instead of just watching it happen.

That excitement doesn't mean I think every problem needs AI thrown at it, and I see a lot of places doing exactly that right now. The truth is, being overzealous with AI can be costly on many fronts, from money spent automating a process that still runs on bad data to tools nobody ends up trusting enough to use. That's a distinction I can make because I spent twenty years working inside the systems and processes underneath these problems, well before AI ever entered the picture. The pattern I keep running into is systems built one at a time, by different teams, at different points, that were never designed to talk to each other. That's usually where things actually break, not inside any one system, but in the seams between them.

I've spent a lot of my career in exactly those seams: data integration, master data management, getting one system's version of the truth to match another's. Not glamorous work, but it's usually the real problem underneath whatever the client thinks the problem is.

I'd rather tell a client they don't need something than sell them a workflow solution that doesn't fit. I also think it's funny how many people are branding themselves as longtime AI experts for a technology that's maybe three years old and still changes every few months. I try not to oversell myself either, if I haven't done something before, I'll say so up front instead of learning quietly on someone else's dime and calling it expertise later.

What I'm building now, Iron Rung Consulting, is that same instinct with a new tool in the mix. I try to figure out where AI actually earns its place, and tell you honestly when it doesn't. Being transparent about it means telling you when the real problem is something else entirely. If that sounds useful, I'd love to hear about what you're working on, and we can talk through what makes sense, whether that's AI, a process fix, or something with your data instead.