IIT Madras alumni-founded startup Discovered Materials has raised $9 million in seed funding to accelerate AI-powered semiconductor materials discovery, with a focus on reducing heat and improving chip efficiency.
As artificial intelligence drives demand for more powerful semiconductors, managing the heat generated by advanced processors has become a significant engineering challenge. Discovered Materials, a deep-tech startup founded by IIT Madras alumni Akash Ramdas and Advaith Sridhar, is developing AI agents to identify new materials that could improve semiconductor performance, thermal management and energy efficiency.
The company announced a $9 million seed funding round in August 2026, led by Lightspeed India Partners, with participation from Y Combinator, Peak XV Partners and angel investors including Paul Graham, Gokul Rajaram and Thariq Shihipar. The funding will support laboratory expansion, team development and the scaling of AI-powered research systems designed to accelerate semiconductor materials discovery.
Key Highlights at a Glance
- Funding: $9 million in seed investment led by Lightspeed India Partners.
- Founders: IIT Madras alumni Akash Ramdas and Advaith Sridhar.
- Core technology: AI agents combined with physics-based simulations and experimental testing.
- Research focus: Advanced semiconductor materials, initially targeting thermal management and heat dissipation.
- Research progress: The startup says it has synthesised and tested thermal interface materials comparable in performance to established industry products.
- Open-source initiative: Material Discovery Bench, a research benchmark evaluating AI models on semiconductor materials-discovery challenges.
The Cost of Heat in Modern Data Centers
Modern computing relies on Graphics Processing Units (GPUs). These chips process large sets of data. As they work, they generate a high amount of thermal energy. Discovered Materials notes that high-performance GPUs handle heat fluxes of about 140 watts per square centimeter. To put this in plain words, that small chip surface gets very hot, very fast.
If this heat stays on the chip, the system must slow itself down. This prevents the silicon parts from melting. In the past, the tech industry fixed this heat issue by building bigger cooling fans. They also used liquid cooling pipes. However, hardware makers now pack more transistors into tiny spaces. As a result, standard cooling methods are not enough to keep the chips safe.
The core problem now lies in the materials. Data centers require new technology to run efficiently. To understand how new server demands push infrastructure limits, you can look at how the central bank builds its secure networks. The new RBI data centre in Bhubaneswar also relies on modern thermal management to run its sovereign cloud operations without interruption.
What Are Thermal Interface Materials?
The industry needs better Thermal Interface Materials (TIMs). These are the specific pastes and pads that sit between the hot silicon chip and the metal cooling hardware. If this material does not transfer heat quickly, the cooling fan cannot do its job. The heat stays trapped, and the server runs slow.
A better TIM allows the chip to run at its top speed for much longer. It also cuts down the massive electricity load needed by data center cooling fans. This specific material problem is the exact issue the startup wants to fix. Finding a better paste can save millions of dollars in electricity costs over a year.
In fact, reducing power costs is vital for the entire tech sector to remain profitable. For context on how operational costs affect different tech business models, consider how quick delivery services shape e-commerce value in India. High overhead costs, like electricity and logistics, define the success of digital platforms.
How AI Changes the Science Lab
Traditional material science is a slow process. A human scientist might write down and test about 20 material ideas per day. The failure rate in physical labs is very high. Mixing chemicals takes time, and most ideas do not work out. Discovered Materials is trying to change this slow timeline.
According to a report from TechCrunch, the startup uses Anthropic's AI models. They placed these models inside a custom software system. These software agents can explore thousands of molecular shapes in the cloud. They run all day, looking for unique bonds that might conduct heat better than current standard pastes.
However, the startup does not blindly trust the AI output. AI models often invent things that are not real. Therefore, the team routes these AI ideas through strict physics software models. These computer tests weed out the bad ideas before the team buys any physical chemicals.
| Research Phase | Traditional Lab Method | Discovered Materials' Approach |
|---|---|---|
| Finding Ideas | ~20 manual guesses per day | Thousands of ideas generated via AI |
| Computer Screening | Months of manual data entry | Automated physics tests |
| Physical Tests | High failure rate in labs | Lab tests only on top passing items |
DeepMind GNoME vs. Vertical AI Focus
It is important to note that Discovered Materials is not the only group using AI for science. Google DeepMind recently created a tool called GNoME. This tool found over two million new crystal structures. However, GNoME is a general tool. It looks for many types of materials, like battery parts or solar cells.
In contrast, Discovered Materials uses a vertical focus. They only look for semiconductor cooling materials. By focusing on just one hardware problem, the team can move much faster. They do not have to filter through thousands of battery materials. This tight focus is what attracts early-stage investors.
The Manufacturing "Valley of Death"
Finding a new material on a computer screen is relatively easy today. Physically making that material at a large global scale is very difficult. In the hardware industry, this gap between theory and mass production is known as the "Valley of Death."
A new thermal paste might move heat perfectly in a small lab test. But if it breaks down over time, or if it costs too much to make, large chipmakers will reject it. The material must remain stable under intense pressure and frequent temperature changes for many years.
The startup has laid out a strict timeline. They aim to have patent-ready commercial materials prepared by early 2027. To hit this target, they will need to pass hard testing cycles. They must prove to massive factory plants that their new compounds are safe for commercial assembly lines.
This is why the recent $9 million seed funding is necessary. Making software is cheap, but renting physical chemistry labs and buying rare chemicals costs a lot of money. This funding allows them to bridge the gap between digital ideas and physical products.
The Open-Source Material Discovery Bench
To prove their AI works, the company launched the Material Discovery Bench. This is an open-source testing tool. It evaluates if AI models can actually solve real material science problems.
The company built this test with input from experts at IBM and Stanford. It acts like a strict lie detector for artificial intelligence. A basic AI model might suggest a material that breaks the laws of physics. The benchmark tests if the proposed material is actually stable and suited for a harsh computer environment.
"The benchmark is intended to assess whether AI systems can contribute meaningfully to materials-discovery tasks rather than merely produce plausible scientific explanations." — Discovered Materials Official Website
This open-source tool is a smart plan. It positions the young startup as a trusted player in the AI science space. It forces other AI companies to prove their claims using the same strict rules.
The Founders and the Silicon Valley Funding
The startup has a strong academic background. Co-founder Akash Ramdas holds a PhD in materials science from Stanford University. He spent years studying electronic devices. Co-founder Advaith Sridhar studied AI at Carnegie Mellon University. He built complex software systems in his past roles.
Their combined skills bridge the gap between software and physical chemistry. This mix of knowledge attracted top venture capital. Reports confirm that Lightspeed India Partners led their massive seed round. The fact that Y Combinator and Peak XV Partners also joined shows trust in the founders.
The company plans to license its successful findings directly to large chipmakers. This would allow the startup to build a business model based on intellectual property. They can collect royalty fees without needing to build their own costly factories.
The Shift in Indian Startups
For India’s growing deep-tech space, this startup represents a big shift. For the last ten years, Indian founders mostly built software apps or consumer delivery services. Now, founders with Indian academic ties are tackling hard physical engineering problems.
Creating new chip materials requires highly specific science skills. It demands large computer servers, expensive labs, and connections to global hardware makers. It is a slow process that requires a lot of patience from investors. Private capital is just now starting to back these bold ideas. You can read more about changing capital trends in an interview with Dr. Amit Goenka on how private credit funds new business growth.
AI can certainly speed up the daily work of research teams. It can help them check strange chemical shapes and pick which tests to run first. However, the final success of Discovered Materials will depend entirely on the physical world. Their new paste must cool down a running server better than the current market leader.
The next major steps for the company will be filing new patents. We must also wait for test results from outside labs to confirm their claims. Until then, the firm remains a very interesting and well-funded science project. You can read more about their early work in a recent Economic Times feature.
Frequently Asked Questions (FAQs)
Quick answers about Discovered Materials, their AI tools, and the chip heat crisis.
What does Discovered Materials do?
Discovered Materials is a tech startup. It uses artificial intelligence and physics tests to find new materials that improve heat removal in computer chips.
Who founded Discovered Materials?
The company was founded by IIT Madras alumni Akash Ramdas, who holds a PhD in materials science from Stanford, and Advaith Sridhar, an AI engineer from Carnegie Mellon University.
How much funding has the startup raised?
The startup raised $9 million in a seed funding round. The investment was led by Lightspeed India Partners, with help from Y Combinator and Peak XV.
What is the Material Discovery Bench?
It is an open-source testing tool created by the startup. It tests whether various AI models can actually solve real material science and chemistry problems.
Why do AI chips need new materials?
Modern AI chips produce high heat, reaching up to 140 watts per square centimeter. Current cooling materials are hitting their limits and need replacement to stop computer failures.
Editorial & Legal Disclaimer: This report is published by News4Bharat for general informational and journalistic purposes only and does not constitute financial, investment, legal, or professional advice. All data, figures, and statements are sourced from public filings, regulatory notifications, and attributed third-party reports believed to be reliable at the time of publication; however, figures marked as approximate, provisional, or reported remain subject to official audit and revision. References to ongoing investigations, disputes, or regulatory actions reflect alleged or preliminary positions unless confirmed by a final judicial or statutory order. Readers are advised to verify details independently with certified professionals or primary regulatory records before making any financial or legal decisions.