Impact
Specific products
Structural foundation
Teams don’t have to spend as much time figuring out which routes are safe and usable.
Organizational capability
Administrative Capacity
Essent has a framework for advancing AI enablement, scaling it up, and connecting it to the community and AI literacy.
About
The energy market no longer demands just cheap electricity. Customers want charging stations, heat pumps, home batteries, and smart contracts—preferably without having to become energy managers themselves. For Essent, this represents a fundamental shift: from a traditional supplier to a partner in smart energy management.
In the article, Roos Anneveldt, Head of IT Technology, deliberately describes Essent as a “first follower.” Not the first to try something, but agile enough to follow quickly as soon as the market shifts. AI operates precisely within that tension. It can accelerate teams’ work and enable new services, but only if the foundation is right. Otherwise, AI tends to develop mainly through isolated initiatives, while the organization actually needs control, reusability, and trust.
“A year ago, I felt a tremendous sense of urgency, but I had little idea of where we needed to go. Now I no longer feel that way. The discussions, analyses, and advice from Mathijs and the team still form the basis for my actions, even in my new role.”
When everyone takes the initiative, direction becomes more important
Within Essent, there was already extensive experimentation with AI in dozens of areas at once, though not always in a coordinated manner. That calls for coordination. How do you ensure that all that energy comes together to form a mature capability?
Roos recognized that moment early on. The leadership supported the move toward more AI, but a practical implementation was lacking. Which building blocks are truly necessary? Where do compliance risks arise? How can teams accelerate without compromising reliability and traceability? As Roos puts it: “There was a lot going on everywhere. We all had the feeling that something was still missing. But what, exactly?”
The challenge wasn’t a lack of ambition. There was plenty of that. The challenge lay in the next step: moving from initiative to foundation. Mathijs van Bree, Principal AI Consultant at Team Rockstars, emphasizes Roos’s pioneering role: “Everyone wants more AI, but often there’s no strategic framework. Roos recognized the urgency, asked the right questions, and created the space to tackle it herself. That kind of leadership is a good fit for an organization that’s truly committed to AI.”
AI enablement to make faster, better decisions
Team Rockstars chose to “understand first, then build.” They began with a broad discovery phase: fifty interviews with approximately forty people, ranging from enterprise architects and product owners to CISOs and teams already working on AI use cases.
That approach provided both clarity and an outside-in perspective. Team Rockstars observes how AI platforms are set up in enterprise organizations, which choices work, and where central teams inadvertently become bottlenecks. That pattern recognition helped Essent connect its own situation to what has been proven to work in the market.
The discovery phase yielded three recommendations: a GenAI platform for the secure use of models and reusable components, an internal AI chat tool as a secure alternative to the proliferation of shadow IT tools, and a community-based approach to adoption. The first two have been implemented; the third has now also been embraced by management.
The key question underlying all the technical decisions—regarding the E.ON AI Gateway, LightLLM, AWS, and building blocks in Python and TypeScript—was always the same: What should be managed centrally to ensure reliability and compliance, and what should be left to the teams to maintain speed?
Mathijs sums it up succinctly: “You want to eliminate repetitive work and tasks where things can go wrong for teams, so they can pick up the pace. At the same time, you have to prevent the central platform team from becoming the bottleneck itself.”
This became clear with the E.ON AI Gateway. While Essent was exploring its own path, it turned out that E.ON was working on a similar solution. Team Rockstars quickly evaluated the three options: build it in-house, choose an alternative, or adopt the E.ON standard. Roos: “Within a few weeks, we had a well-reasoned analysis. That allowed us to make a strategic decision that would otherwise have taken us much longer.”
That’s where AI enablement comes into its own: in the ability to make better decisions faster.
From Individual Initiatives to an Organized Capability
The impact isn’t reflected in a single spectacular metric; AI enablement is still too early in its development for that. The value is more fundamental: Essent can make more targeted choices, scale up more safely, and continue building more independently. The organization has a clearer understanding of which AI capabilities need to be centrally managed, and where teams actually need leeway to maintain their pace.
- Specific products: a secure internal AI chat tool and a GenAI platform for reusable, scalable applications.
- Structural foundation: Teams don’t have to spend as much time figuring out for themselves which routes are safe and usable.
- Organizational capability: knowledge, decisions, and responsibilities related to AI are less dependent on isolated initiatives or individual people.
- Organizational Capacity: Essent has a framework for advancing AI enablement, scaling it up, and linking it to community engagement and AI literacy.
Roos Anneveldt, Head of IT Technology: “Mathijs and the team’s analyses and recommendations still form the basis for how I interact with the rest of the organization.”
Collaboration
The collaboration worked because Team Rockstars didn’t come in with a blank slate. Roos explicitly points this out: “They know Essent, understand our market context, and can therefore respond much more quickly to actual demand.”
Not by taking over the process, but by contributing knowledge, context, and outside-in experience at the moment decisions are being made. For Roos, therefore, the partnership has no fixed end date. The conversation continues—about AI literacy, competency development, and what Essent can learn from other organizations. Not driven by sales pressure, but by a substantive relationship that naturally leads to the next step.
The question isn’t whether people will experiment. They’re going to do it anyway. The real question is: Can you build a foundation that accelerates teams while maintaining reliability and traceability? In summary, Mathijs says: “We ended up with two products that are embedded in existing teams—not as an external project, but as part of the organization. That’s the goal of enablement: for your presence to leave behind something that grows without you.”
“What sets Team Rockstars apart is that they get up to speed incredibly quickly. They know Essent, understand our market context, and can therefore immediately address the actual needs.”
Want to know more?
Please fill out the form below and a colleague will contact you as soon as possible.
