Discovered Materials uses swarms of AI agents for a chip-materials search that takes more than ten years of lab work. Agents propose materials and synthesis recipes, while machine-learning tools compute their stability, thermal, dielectric and mechanical properties. The team makes best-effort attempts to validate candidates in its lab. Advaith has a master's in AI from Carnegie Mellon and was the founding applied scientist at Persona AI, acquired by Luma Labs, where he built long-horizon autonomous agents for major telecom and crypto companies. Discovered Materials (YC P26) raised a $9M seed in August 2026 led by Lightspeed India Partners. It has released more than 500 computationally generated, previously unknown materials and Material Discovery Bench, its open benchmark for agentic materials discovery.
Episode featuring Advaith Sridhar will be out soon.
Advaith Sridhar is a co-founder of Discovered Materials, which builds AI scientists that discover materials for the semiconductor industry. He holds a master's in AI from Carnegie Mellon and was the founding applied scientist at Persona AI, acquired by Luma Labs, where he built long-horizon autonomous agents.
A San Francisco company in Y Combinator's Spring 2026 batch (YC P26), building AI agents that find new materials for semiconductor chips, aimed at data centers and fabs. Finding a novel material today takes more than ten years of lab work. The company aims to compress that to months.
Agents propose materials and synthesis recipes, while machine-learning tools compute stability, thermal conductivity, dielectric properties and mechanical properties. The company says it makes best-effort attempts to validate candidates experimentally; its published benchmark found only one plausible synthesis route among more than 500 computationally generated materials.
The company's open benchmark for agentic materials discovery: a long-horizon, open-ended research task where AI models search for new materials used in chips. It was built with input from researchers at IBM, imec, Stanford and Cambridge.
A $9M seed round announced in August 2026, led by Lightspeed India Partners with participation from Y Combinator and Peak XV Partners, and angels including Paul Graham, Gokul Rajaram and Thariq Shihipar.
Akash Ramdas, who did his MS, PhD and post-doc at Stanford in materials science and engineering. The materials he found for nanoscale interconnects during his PhD have been adopted into the roadmaps of Intel and TSMC.
