Builders of AI: Meet Alexandru

One of the core challenges marketing teams face is not a lack of ideas. It is knowing where to invest. Alexandru tackled this by building a custom investment distribution model. At a high level, it helps answer a critical question: how should a marketing budget be allocated across channels and regions to maximize return? The model does not just spread budget evenly or follow past patterns. It analyzes performance signals and recommends where investment will have the greatest impact. But understanding where to spend is only part of the equation.

It is easy to see when two things move together. It is much harder to prove that one actually caused the other. That distinction matters in marketing. To address it, Alexandru developed a second model focused on causal analysis. Instead of simply tracking performance trends, it evaluates whether a campaign directly drove an uplift in results. “We’ve developed a custom model designed to analyze campaign performance causally, not just correlationally,” he explains. “That means we can show if a campaign actually caused an improvement or not. We can also use it for forecasting to help stakeholders plan future investments.” This shifts conversations from assumptions to evidence. It also gives teams the confidence to double down on what works and rethink what does not.

Beyond modeling, Alexandru has also built tailored marketing agents. These are automated systems that support campaign optimization, audience targeting and experimentation. They do more than report on performance. They actively help improve it. This kind of work reflects a broader shift in how AI is being applied. It is no longer just about insights. It is about enabling action.

Alexandru’s path into Publicis Sapient came through an acquisition, which gave him a unique perspective on what feels different. Two things stand out to him. “It’s the variety and the culture,” he says. Even while embedded with a single client, he is not working in isolation. Teams regularly share demos, experiments and new approaches across the organization. That visibility matters. It exposes engineers to different use cases, tools and ways of solving problems. It also creates an environment where learning is shared rather than siloed. “The environment is collaborative rather than competitive,” he explains. “People are working toward common goals and actively helping one another learn.”

Like many working in AI, Alexandru has seen the speed of development change dramatically. “What might have taken a week to prototype can now be done in a day with modern AI tools,” he says. “If you know what to ask, what to look for and what results to expect.” That speed opens up new possibilities, but it also introduces risk. Faster does not always mean better.
“Not all shortcuts are safe or valid,” he adds. “Critical thinking is still essential when working with these tools.” It is a reminder that while technology evolves quickly, judgment remains a constant requirement.

Technical skills matter. Alexandru is quick to acknowledge that. Experience with Python, databases and machine learning libraries forms the foundation of the work. But he does not see those as the defining factors for success. Those skills can be taught. What makes the difference is how someone approaches the work. “I believe the qualities that help someone thrive go beyond code,” he says. “Curiosity, eagerness to learn, being open to sharing knowledge and having a team-first mindset.” In a field that changes as quickly as AI, those traits are what allow people to adapt, grow and keep moving forward.

Alexandru’s work is not about building models for the sake of it. It is about making complex systems understandable and actionable. When marketing teams can clearly see where to invest, understand what is driving results and plan with confidence, the impact goes beyond a single campaign. It changes how decisions get made. And that is where the work starts to matter.