Two decades shipping production systems that have to actually work

I spent two decades at Trimble, growing from a software engineer writing code to a Senior Engineering Director leading globally distributed teams of AI engineers, data scientists, and robotics specialists. That trajectory gave me something that's hard to replicate: deep understanding of every layer of a technology organization, from the code that runs on embedded devices to the board-level conversations about where to invest next.

At Trimble, I built and led teams developing next-generation products for the construction and geospatial industries. The work was technically demanding in ways that pure software rarely is. We were building AI systems that needed to operate in the physical world: on construction sites, in agricultural fields, in mining operations. That meant dealing with real-time sensor data from GNSS receivers, total stations, laser scanners, and drones. It meant building edge AI systems that had to make decisions with limited connectivity and strict latency requirements. It meant developing machine learning models that processed spatial data at scales that most software engineers never encounter.

Some of the work I'm most proud of happened at the intersection of technologies that don't usually sit together. I led early mixed-reality integrations with Google, developing spatial computing products that overlaid digital information onto the physical world through Trimble SiteVision. I built analytics platforms that processed real-time IoT data from thousands of connected devices to deliver actionable insights to field operators and project managers. I developed computer vision systems for construction monitoring and autonomous machine guidance.

I also served on the board of Trimble Europe B.V., which gave me a different perspective on technology leadership, understanding how technical decisions connect to business strategy, governance, and the long-term trajectory of a company operating across dozens of countries.

Today I lead AI initiatives at Orange Logic, where I'm developing intelligent agents, context graphs, and machine learning systems that transform how organizations manage and understand their digital content. The work involves building AI systems that can reason about complex relationships between digital assets, automate classification and discovery workflows, and provide contextual intelligence that makes large content libraries genuinely useful rather than just searchable.

This role keeps me current in a way that matters. I'm not advising based on what AI could do in theory. I'm building production AI systems every day. I understand the practical realities of prompt engineering, model evaluation, RAG architectures, agent frameworks, and the operational complexity of deploying AI in enterprise environments. When I advise clients on AI product strategy, I'm drawing on work I did this week, not work I did five years ago.

Looking back across two decades, the consistent thread is this: I've always worked at the intersection of technical depth and product thinking. The best products I've built were not the most technically sophisticated. They were the ones where deep technical capability was aimed precisely at a real customer problem. That's the lens I bring to every engagement. Not "what's the most advanced thing we can build?" but "what's the most valuable thing we can ship, and what's the fastest credible path to get there?"