Insider Brief
- Mariana Minerals raised $310 million in Series B funding led by Khosla Ventures, bringing its total parent- and project-level capital raised to about $400 million.
- The company will use the capital to expand its critical-minerals projects and advance MarianaOS, an AI platform that links project development, mining operations and refining.
- MarianaOS includes CapitalProjectOS for construction and procurement, MineOS for coordinating mine operations and PlantOS for using sensors and reinforcement learning to optimize refining.
Mariana Minerals announced raising $310 million in Series B funding to expand its critical-minerals projects and further develop an AI platform designed to automate mine construction, extraction and refining.
According to the San Francisco-based mining company, Khosla Ventures led the round, with continued participation from Andreessen Horowitz and Breakthrough Energy Ventures. New investors included Greenoaks, Halo Fund, Pax Ventures, StepStone Group, BHP Ventures, Washington Harbour Partners, Greycroft, General Innovation Capital Partners, Mitsubishi Corp., In-Q-Tel and Earthshot Ventures.
The financing brings Mariana’s total parent- and project-level capital raised to about $400 million, according to the company.
Mariana said it will use the capital to expand its portfolio of mining and refining projects and advance MarianaOS, its software system for managing operations from initial development through mineral processing.
The company is targeting rising demand for critical minerals used in AI infrastructure, power grids, electrification and aerospace. It said new production capacity has become harder and more expensive to build as ore grades decline, deposits become deeper and mineral processing grows more complex.
MarianaOS is already being used at the company’s Copper One and Lithium One sites. The platform includes three systems focused on project construction, mine operations and refining.
MarianaOS Platform
CapitalProjectOS
CapitalProjectOS brings engineering documents, equipment data, procurement, schedules and costs into one system.
Mariana said the platform uses AI agents to automate administrative and coordination work across engineering, procurement and construction. The company claims the system can cut project-development time and cost by reducing manual workflows and disconnected data.
MineOS
MineOS coordinates decisions across geology, drilling, blasting, vehicle fleets and processing.
The system combines equipment and operating data with physics-based simulations and reinforcement learning, a form of AI that improves by testing actions and evaluating the results. Mariana said it is designed to reduce manual coordination and make mine operations more consistent across shifts.
PlantOS
PlantOS applies sensors, digital simulations and reinforcement learning to mineral-processing plants and refineries.
The software creates models of plant operations and trains AI systems to adjust processes based on sensor feedback. Mariana said it is using the platform across seven operations at Copper One and Lithium One, including lithium extraction, reverse osmosis, bioleaching, solvent extraction and electrowinning.
Mariana said it compressed the acquisition, permitting and restart timeline at Copper One from years to months. At Lithium One, the company expects to bring an automated greenfield refinery online less than three years after Mariana’s founding.
