Startup Discovered Materials raised $9 million to search for materials for cooler chips using AI agents
Startup Discovered Materials closed a $9 million seed round from Lightspeed India Partners after completing the Y Combinator accelerator. The company uses AI agents with models from Anthropic to search for materials for cooler and more efficient semiconductors and has released the Material Discovery Bench tool.
Startup Discovered Materials, founded by Advaith Sridhar and Akash Ramdas, closed a $9 million seed funding round. The round was led by Lightspeed India Partners, with participation from Peak XV Partners and angel investors Paul Graham, Gokul Rajaram and Thariq Shihipar. The company was established after completing the Y Combinator accelerator.
According to Discovered Materials, the founders built a software pipeline that uses models from Anthropic in a modified harness to generate proposals for new materials; the proposed candidates are then validated using proprietary physics simulation models. The goal is to find materials that enable cooler and more efficient semiconductors, which should address overheating in chips handling AI workloads in datacenters. According to co-founder Advaith Sridhar, Ramdas could perform roughly 20 estimates a day during his PhD, while agents running continuously in the cloud can now perform thousands.
The company published examples of hundreds of newly proposed materials and released a tool called Material Discovery Bench, designed to track the performance of frontier models on this type of task. According to the company, it has already found materials with properties matching those of materials used by major chip manufacturers, but it is not sharing details yet. According to the source, MatNex, SandboxAQ and CuspAI are pursuing a similar direction; Discovered Materials distinguishes itself through its narrow focus on the thermal properties of semiconductor materials. The founders hope to find materials suitable for patenting and licensing to chip manufacturers within a year. The source also states that no AI-discovered materials or drugs have yet reached commercial deployment at scale.
Why it matters
It demonstrates a concrete application of AI agents beyond software - in materials research for semiconductors, aiming to reduce heat losses in chips and the associated energy consumption for datacenter cooling. According to the source, however, this is still at an early stage: the company itself states that no AI-discovered materials have yet reached commercial deployment, so the practical impact depends on further laboratory validation and potential patenting and licensing to chip manufacturers.
Relevant practical impact
What this means
For a business
For chip manufacturers and companies in AI infrastructure, this signals growing investor interest in AI agents deployed in materials research as a way to reduce datacenter cooling costs; the company currently offers only early candidate materials, with no commercial results reported.
DevelopmentCheck the original
Event sources
only one source so far · 1 publisher, 1 independent. We count feeds from the same owner only once.