CloudX AI Labs designed the AI architectural foundations for JEN, Mainspring Energy's agentic diagnostic system for linear generators. In three weeks, the engagement delivered a current-state architecture map, a future-state agentic design, and a production-ready implementation blueprint.
Mainspring Energy is a California-based linear generator manufacturer redefining how local power is delivered to capacity-constrained companies and communities. Their solutions serve utilities, data centers, enterprise, and industrial clients across key markets, with over 450 MW in field operations and late-stage development for Fortune 500 organizations.
Recognizing the opportunity to accelerate diagnostic operations with AI, Mainspring launched JEN: an innovation initiative to explore how AI could support fault diagnosis across its generator fleet. With an initial proof of concept successfully validated, Mainspring brought CloudX in to design the architecture that would take JEN to production scale across hundreds of distinct fault scenarios.
"CloudX's expertise in our architecture, combined with a strong focus on evaluation and observability, significantly reduced our time to get started. We now have a clearer path toward autonomous operation, and collaborating with the team was a great experience."
Senior VP of Engineering at Mainspring Energy
The gap between a validated proof of concept and a production-ready AI system is where most enterprise initiatives stall. It requires solving problems that do not exist at pilot scale: cost efficiency per inference cycle, context management across complex integrations, workflow automation, evaluation frameworks, and a long-term architectural direction that can evolve with the business.
For JEN, that transition meant moving from a single validated workflow to an architecture capable of reasoning across hundreds of fault scenarios with consistency and economic viability. Mainspring engaged CloudX precisely at that inflection point.
Mainspring needed architectural clarity and a strategic foundation before committing to a full production investment.
The engagement was structured as an AI Labs Large project: a fixed-scope, senior-led exploration designed for complex and ambiguous AI challenges that require broad discovery to define the precise path forward.
CloudX deployed a senior Fast Impact Team with Erik Davidsson, Head of AI, leading a one-week on-site immersion at Mainspring's facilities. Working side by side with Mainspring's engineering team, the team executed:
Together, the teams defined a structured agentic approach that reframed JEN as a scalable AI system: coordinated workflows with explicit evaluation and observability, and token economics built into the architecture.
“The goal of this first stage was not about blindly defining an architecture or scaling a PoC, it was on generating a deep understanding of the domain expert's workflows. We wanted a laser-focused solution with real impact, and the first step is always knowing where to point that laser. The JEN PoC showed us the right path but we needed higher precision.”
Head of AI at CloudX
Mainspring's engineering team received a clear, rigorous, and well-documented target architecture, enabling the engineering team to move forward with confidence and precision.
As the architectural design took shape, Mainspring identified opportunities that had not been visible at the outset. CloudX maintained speed and rigor throughout, demonstrating in practice what Agentic Innovation produces: a structured discovery process that expands what is possible without losing momentum.
Documented achievements:
The strongest signal of a successful engagement is a client ready to move forward with confidence. That is where this partnership left Mainspring, and where every CloudX engagement is designed to end.