Emerald AI, a Washington, DC-based startup that turns data centers into flexible grid assets, has closed a $150M Series A at a $1.05B valuation. Energize Capital and DCVC co-led the round, which drew an unusually broad investor list including Nvidia, Samsung Ventures, Siemens, Aramco Ventures, Salesforce Ventures, GE Vernova, RWE, JERA Ventures, In-Q-Tel, John Doerr, Tom Steyer, Lowercarbon Capital and Energy Impact Partners.
The company is led by Varun Sivaram, and its flagship product, Emerald Conductor, coordinates AI compute workloads with onsite generation and storage so a facility can trim its power draw during grid emergencies while keeping model training running. The software effectively lets a data center behave like a demand-response asset. Buyers span AI companies, data center owners and utilities.
The thesis is that data centers are becoming some of the largest new electricity loads on the grid, and that they can behave as responsive assets rather than fixed consumers. As AI compute scales and operators chase scarce power, the ability to shed or shift load on demand is taking on real economic value.
The company says it will use the capital to expand operations and development efforts around its orchestration platform.
