Funding · AI & Deep Tech

Discovered Materials raises $9m seed round led by lightspeed india partners

By Startup Enthusiast · 10 August 2026

What they do

The company uses AI agents to discover thermally conductive dielectric materials for semiconductor chips

Lightspeed India Partners led the $9 million seed round with participation from Y Combinator and Peak XV Partners

Founders are Stanford PhDs who previously worked on agent research at Persona AI and Luma Labs

Discovered Materials has raised $9 million in a seed round of funding.
Lightspeed India Partners led the investment round as the primary backer.
Y Combinator and Peak XV Partners participated in this financing round.
Notable angels including Paul Graham and Gokul Rajaram also invested.
The startup is a graduate of the YC W26 Demo Day cohort.
It aims to solve heat management challenges in advanced semiconductor packaging.
The team plans to use funds to expand its lab and research capabilities.
The company released an open-source benchmark called the Material Discovery Bench.

Heat dissipation is a critical bottleneck for high-performance GPU computing.
Current GPUs handle approximately 140 watts per square centimeter of heat flux.
The founders have deep academic credentials from Stanford and IIT Madras.
The startup competes with firms like MatNex and SandboxAQ in the AI materials space.

The firm focuses on discovering thermally conductive dielectric materials for chips.
It utilizes cloud-based autonomous AI agents powered by Anthropic models.
These agents run physics simulations to screen thousands of material candidates.
The business model involves patenting promising candidates for licensing to chipmakers.
It targets the growing market for 3D packaging thermal management solutions.
The Material Discovery Bench benchmarks AI capability in computational materials science.
Frontier models identified over 500 previously unknown computational material candidates.
The team simulated and tested materials matching major chemical company products quickly.
Collaborators include experts from IBM, IMEC, and Cambridge University.

Product: Cloud-based AI agents for automated materials discovery and optimization for semiconductor thermal management
Customers: Semiconductor manufacturers and chipmakers requiring thermal dissipation solutions for advanced processors and AI chips
How it makes money: B2B deep tech platform providing materials research and development services using AI automation
What sets it apart: Uses autonomous AI agents running thousands of virtual material simulations daily to accelerate discovery from decades to months
Operations: AI-powered materials discovery combining custom model harnesses with frontier AI, physics simulations and experimental validation