Discovered Materials closed a $9 million seed round led by Lightspeed India Partners, with participation from Peak XV Partners and angel investors including Paul Graham, following the startup’s emergence from Y Combinator. The company aims to deploy swarms of AI agents to discover new materials capable of building more thermally efficient integrated circuits, addressing the heat and energy demands driving up data center electricity consumption.
Founders Advaith Sridhar and Akash Ramdas, drawing on Ramdas’s doctoral research in materials science at Stanford and Sridhar’s background working on AI agents, built a software pipeline combining Anthropic models with custom-trained physics models to generate and simulate candidate materials. Sridhar told TechCrunch that the system now runs thousands of exploratory guesses daily, compared to roughly twenty a day during Ramdas’s PhD work. The company released examples of newly identified materials alongside a benchmark tool for tracking how AI models perform at the task.
Lightspeed partner Hemant Mohapatra, who led the round, said identifying viable materials resembled solving a complex convergence problem, since candidates must satisfy multiple properties simultaneously to prove useful. He noted that filtering and synthesizing promising candidates, rather than generating them, remains the primary bottleneck facing the field.
Featured image: Credit: Discovered Materials
