Alinta Energy spins up AI system for executive insights
Alinta Energy has built a multi-agent AI system that provides executives with insights and confidence scores to aid decision-making.
Alinta Energy, an Australian electricity generator and retailer, has developed a multi-agent artificial intelligence system called the Alinta Intelligent Strategic Engine (AISE) to provide executives with data-driven insights and confidence assessments.
The system was born out of a collaboration between the CEO and CIO, who wanted to leverage the company's investment in its Databricks platform over the past six to seven years. Built over four weeks with approximately 70 person-days in partnership with Databricks, AISE integrates 500 structured data metrics and over 9,000 documents that the board and executive team rely on, including board papers, strategy decks, and market research.
A supervising agent determines the nature of each executive query and routes it to the appropriate subsystem. Results are returned with a confidence score that guides executives on how much weight to place on the response. The scores indicate whether the executive can make a decision based on the insight, use it for directional thinking, or treat it as generally correct but not yet reliable enough for critical decisions. Every insight is linked back to the specific document or metric from which it was derived.
AISE is now in production, with ongoing work to build AI agents that interact with metrics and documents from each business unit.