Solar panels installed across a utility-scale renewable energy facility.

AI capability embedded to cut operational costs across an off-grid solar network

Specialist AI and data science engineers extended an EU energy operator's team — building predictive energy models and automated site reconfiguration logic across a growing off-grid solar portfolio.

SECTOR
Renewable Energy
CAPABILITY
Team Augmentation
REGION
Europe

THE CLIENT

A European renewable energy company operating a network of off-grid solar sites equipped with battery storage and diesel generator backup. The company's operational mandate was precise: reduce fuel consumption, cut carbon output, and improve cost efficiency across a portfolio designed to grow. The business case for this mandate was clear. The internal capability to deliver it through AI and machine learning was not available.

THE CHALLENGE

The operational challenge was a forecasting and automation gap. Diesel gensets were operating beyond the point at which solar and battery capacity would have been sufficient to replace them, because the team had no reliable way to forecast when that threshold would be met. 

Battery storage was regularly reaching capacity, meaning solar energy that could have been captured and stored was instead lost. External forecasting data was available but the team lacked the data engineering skills to integrate and operationalize it. Hiring permanent AI and data science talent locally would have been slow, expensive, and disproportionate for a use case with a well-defined scope.

WHAT WE BUILT

Primero Group placed a specialist AI and data science team alongside the client's existing operations engineers, extending the team's capability without displacing its operational knowledge. The work covered three layers: integration of third-party weather and solar forecasting data into the client's site infrastructure; development and validation of machine learning models predicting daily energy production per site; and design of automated daily site reconfiguration logic that determined optimal genset and battery deployment based on forecast output. Knowledge transfer was built into the engagement structure from the outset — the goal was for the internal team to be able to operate and extend the models independently.

20%
Reduction in operational costs
200
Sites approved for rollout
Improved ESG
Measurable, attributable carbon reduction

WHAT IT SHOWS

This engagement demonstrates our ability to deploy senior AI and data science capability on-demand, at the speed that market conditions require, into sectors where that expertise is structurally scarce. The approval for 200-site rollout is an indicator of validated delivery. The client assessed the work against operational outcomes before committing to program-wide expansion.