From VINs to Value: How Daimler Trucks Is Building the Future of AI-Driven Manufacturing

Edgar Gallo, Chief Data Officer at Daimler Trucks North America (DTNA), shares how his team is transforming truck manufacturing through AI agents—and why metadata matters more than ever.

When I joined DTNA, I quickly realized something most people overlook: nearly everything you buy was moved by a truck. We’re part of the backbone of the global economy—and as we innovate, we enable others to innovate too.

Today, we’re leading our next transformation: one where AI agents don’t just analyze data but actively shape decisions. At Alation’s revAlation conference, I shared lessons from that journey—and how metadata made it possible.

From governance to activation

Our work with Alation began in 2019 with a clear goal: catch up with our data. We needed to answer basic questions—Where is our data? What’s in it? Who’s using it?

Alation quickly became the foundation for governance and compliance—powering GDPR and CCPA initiatives, supporting finance, and strengthening transparency across the business. But we soon realized that governance alone wasn’t enough.

Over time, we realized the real shift wasn’t from “data unknown” to “data documented”—it was from governance to activation. As AI capabilities accelerated, our 900-person data team was still asking the same question: “Where’s my data?” That convergence—AI technology meeting organizational readiness—was the turning point. Metadata couldn’t just be a catalog to admire. It had to become the engine that drove action.

The payoff is clear: an end-to-end data architecture that can unlock new service opportunities, predict maintenance needs, and strengthen customer relationships in ways we couldn’t imagine before. And none of it would be possible without the foundation we’ve built with Alation.

Beating the bullwhip: AI for supply chain resilience

Supply chains are unforgiving. The “bullwhip effect”—when small demand shifts cause large supply swings—can cripple operations. That’s why we’re developing AI agents that act as early-warning systems, alerting teams to potential disruptions before they escalate.

Working with Numbers Station, we created “vertical” agents that think like supply chain planners and “horizontal” agents that execute tasks. These agents detect risks, monitor inventory, and surface insights that help planners act more quickly.

The goal is not only to look down the rest of the "Pareto" while looking at delays, but also to shift the focus on the team member from digging through data to partnering with our vendor on a resolution plan and overseeing its execution. 

Why AI won’t take your job, but it will change it

AI doesn’t replace people—it amplifies them. Instead of spending hours on manual monitoring, employees train agents, validate results, and improve processes. That shift creates room for more strategic, creative work.

Trust is key. As teams see agents reflect their expertise accurately, confidence builds. Over time, humans and AI form a cycle of continuous improvement: AI handles repetitive analysis, humans focus on insight and strategy, and together they deliver smarter outcomes.

The result? Stronger products, happier customers, and fewer costly disruptions.

The modern CDO: Bridging governance and innovation

Being a Chief Data Officer often feels like partaking in a three-ring circus—balancing business requirements, IT possibilities, and Legal & Compliance frameworks. But success isn’t about how much data you’ve cataloged. It’s about impact: Is governance driving better decisions and faster outcomes?

That’s our focus at DTNA. A data catalog should not only store information—it should be a learning operation with the building blocks (metadata) to act on it. Partnerships across the business are key: when directors and analysts share how they work, that expertise gets encoded into our agents, freeing teams to focus on strategy rather than oversight.

That’s the essence of modern data leadership—ensuring data isn’t just well-documented, but is a source for a data-product-oriented mindset. 

No metadata, no AI

For me, the turning point came with a simple realization: no metadata, no AI.

AI can only be as good as the data—and metadata—that powers it. Large language models and algorithms may grab headlines, but without high-quality, well-documented data, they’re little more than noise. Metadata is what makes AI agents trustworthy, accurate, and valuable.

That’s why my message to data leaders is twofold. If you’ve already done the hard work of documentation and governance—put it to use. Take the catalog, the policies, the lineage, and turn them into action. On the other hand, if you’re all-in on AI but haven’t built that foundation, you need to pause and do a bit more groundwork. Skipping governance will only set you up for failure.

AI topics and capabilities move at incredible speeds. However, when my team began evaluating new technologies like Numbers Station against copilots and platforms such as Snowflake Cortex, the real breakthrough wasn’t just in comparing features. It was in understanding that success doesn’t come from deploying yet another agent. It comes from making processes more robust, repeatable, and high-quality, leveraging the MCP that Numbers Station demonstrated for us. At the orchestrator level is where we can replicate the knowledge and intuition of the human, to free up their capacity and availability.

Automation can streamline tasks, but it doesn’t remove the responsibility to ensure process quality and continuous improvement. The same applies to AI. If the underlying process is weak, no agent can fix it. If it’s strong, AI can scale it, accelerate it, and make it repeatable.

That’s the real takeaway: governance isn’t a box to check. It’s the launchpad. Metadata isn’t an afterthought—it’s the difference between AI that looks impressive and AI that actually delivers business value.

From cataloging to action: The data quality paradigm shift

People are busy—they don’t have more hours in the day to add new tools or processes. That’s why our approach isn’t to ask for more effort, but to offer a simpler path: give a quick thumbs up, let the agents take over, and let the data quality assessments do the heavy lifting. Once people engage, they quickly see the payoff: from that point forward, the value of metadata becomes inseparable from the value of the data itself.

That’s the shift in paradigm — their contributions become highly valuable and sought-after advice from internal consumers. Consistent use creates trust, which drives adoption, which fuels better outcomes. What used to be the “extra work” of governance now becomes the engine for AI-powered impact.

I’ve seen this evolution firsthand. Early in my career, as a data scientist, the big question was: “Who in IT do I call to get my data?” Later, as maturity grew, it became: “What table do I query to access the data?” But where I see the real future is in semantic thinking. Imagine an environment where instead of worrying about tables and joins, you simply ask: “What chunks of data do I combine to create the semantic layer that solves my business problem?” 

That’s what we’re building toward—a world where metadata becomes active, where you don’t just manage data, you converse with it.

The road ahead

At DTNA, our collaboration with Alation is helping us build that future. We’re transforming governance into active, trusted, and repeatable decision-making. 

The future isn’t about fearing disruption—it’s about welcoming the new kinds of work AI enables. Because ultimately, metadata isn’t documentation—it’s transformation.

Curious to see the solution in action? Book a demo with us today.