Damao Technology Raises $27.5 Million to Scale AI Models for the Power Grid

The Chinese startup builds what it calls Energy MaaS, a fully self-built energy large language model platform, and layers AI-powered virtual power plant applications on top of that foundation, positioning generative AI directly inside the infrastructure that manages and dispatches distributed electricity rather than treating energy as a downstream application of a general-purpose model.

Damao Technology closed a Series B round worth approximately $27.5 million, following an earlier Series A+ round of nearly 100 million yuan, roughly $14 million, meaning this new round is close to double the size of the company's prior raise. The company has already established working relationships with a notable roster of major Chinese enterprises, including SenseTime, CATL, Sugon and Fujian Big Data Group, giving it commercial validation from established technology and battery manufacturing players well before this round closed.

Why an Energy-Specific Model Is the Differentiated Bet

Most generative AI applications in the energy sector so far have applied general-purpose models to energy use cases as one vertical among many. Damao's decision to build its own foundational energy model from the ground up, rather than fine-tuning an existing general model, is a materially more capital-intensive bet, but one the company is wagering will produce meaningfully better performance on the specific technical problem of coordinating distributed power generation, storage and consumption across a grid in near real time, a domain with its own physics, regulatory constraints and failure modes that a general-purpose model was never trained to understand deeply. The virtual power plant application layered on top, which aggregates and coordinates distributed energy resources like batteries and generation assets to act as a single dispatchable unit, is where that domain-specific model actually gets tested against real grid operating conditions.

What This Means for Founders

Damao's raise is a useful signal for any founder building AI applications in a physical infrastructure category with its own deep domain complexity. The decision to build a proprietary foundational model specifically for energy, rather than a thinner application layer on top of a general-purpose model, is a higher-cost, higher-risk path, but the roughly doubled round size from Series A+ to Series B, alongside commercial relationships with established players like CATL and SenseTime, suggests investors and enterprise partners are rewarding that deeper technical commitment in a sector where getting the underlying physics wrong has real operational consequences. A founder weighing a similar build decision in a comparably complex regulated or physical infrastructure category should treat early commercial validation from established industry players, not just funding size alone, as the more meaningful signal that a domain-specific model investment is paying off.

Founders building climate tech, energy infrastructure, or applied AI ventures looking for investors with an active thesis in the space can browse climate tech investors on AngelLinx, or explore active capital on AngelLinx @ angellinx.ai/register.

AngelLinx Intelligence | angellinx.ai