In an interview with e27, Cosmos Innovation Co-Founder and CTO Joel Li explains that solar panel installation has increased significantly in recent years due to growing levels of commitment to sustainability. However, the silicon used in most mainstream solar cells is quickly reaching its efficiency limit due to technological limitations.
This means that the demand has made it necessary for producers to explore alternative technologies for solar cells that can generate greater amounts of energy.
“The most promising technology in the industry today is perovskite silicon tandem … However, commercialising perovskite silicon tandem solar cells can be challenging. Highly complex optimization is typically needed across multiple factors during the process development journey, from R&D to mass production,” he says.
Cosmos Innovation is an AI-focused company developing advanced solar technology with the goal of increased efficiency. Their unique approach leverages AI to optimize the last-mile process challenges in commercializing perovskite silicon tandem solar cells.
Their AI platform, Mobius, uses machine learning algorithms to discover the optimal material, process, and architecture combinations for the most efficient solar cells, reducing time and cost compared to traditional trial-and-error methods.
Also Read: What startups need to know about Claims Code, the new rulebook for making credible climate claims
“Since our inception, Cosmos Innovation has built and continuously improved upon Mobius, supported by leading global investors and minds in AI. This has enabled us to massively accelerate the process development journey from lab to mass production for the solar or semiconductor industry, enabling the building of new devices at a fraction of the time and cost it would take using conventional methods,” Li says.
“The effectiveness and viability of Mobius has been demonstrated through our work with leading companies that have achieved outstanding results. We have been able to shorten recipe development timelines by as much as ten times, and significantly improved target performance metrics.”
In 2022, Cosmos Innovation established an AI-centric manufacturing facility dedicated to advanced solar technology. The company said that within a year, it recorded “remarkable” advancements that led to the significant reduction of the R&D process, which typically spans years, to just a few months.
It relies on Google Cloud’s infrastructure for its AI computational, networking, and storage requirements and aims to expand the utilization of Google Cloud as the business expands.
As a participant in the Google for Startups Accelerator: AI First program, Li says that the support that the company receives has been “incredible.”
“These industry-leading capabilities and expertise accelerated our development of key features and enhancements for Mobius. The program team was also prompt in organizing mentoring sessions, and the mentors provided insightful advice that helped us fully realize the potential of our ideas.”
Also Read: What is left behind in our conversation on climate change
Beyond the academy
Cosmos Innovation was jointly established by Li and Vijay Chandrasekhar. The two co-founders boast unique academic backgrounds.
Chandrasekhar earned his Ph.D. from Stanford University and BS & MS degrees from Carnegie Mellon University, both supported by scholarships from the Agency for Science, Technology and Research (A*STAR) and the National Research Foundation (NRF).
Before co-founding the company, he led the AI effort at the Institute for Infocomm Research, A*STAR, managing a team of over 150 AI specialists and overseeing a portfolio of more than 50 AI projects across more than 10 domains, with a significant focus on semiconductors. He has also held positions at Google Research and Nokia Research Center.
Li also received his Ph.D., Ph.D. Minor, and M.S. degrees from Stanford University, again under the National Research Foundation (NRF) Scholarship. After graduating from Stanford, he served as Group Head at the Solar Energy Research Institute of Singapore, leading a team in developing commercial solar cell technologies.
Under his leadership, the team developed an award-winning process technology, which was recognized as one of the Top 25 PV Manufacturing Innovations and received strong industry interest.
Li has also served as an AI team lead at A*STAR. In addition to his current role at Cosmos Innovation, he also serves as an Adjunct Assistant Professor at the National University of Singapore (NUS).
Also Read: Investing in climate tech: Why investors should focus on impactful, low-hanging fruits
“Powered by our AI technology, we have achieved a near vertical learning curve for our next-gen solar technology and are rapidly catching up with the leading companies in this space despite having spent a fraction of the time and cost typically required by traditional industry players,” Li says.
“Cosmos Innovation is today among only a few companies in the world with AI process optimization technology designed for the solar and semiconductor industry, making us a pioneer and leading innovator in the field.”
When asked about the company’s business model, Li explains that the company specializes in cutting-edge solar panel technology aimed at supplying these innovative solutions to both downstream residential and utility-scale companies. Its strategic focus is primarily on businesses offering residential rooftop solar solutions and selected utility-scale customers.
“In our initial stages, we plan to establish a strong presence in the US and Singapore markets. By capitalizing on the US’s dominance in the solar energy sector and Singapore’s increasing emphasis on sustainability, as outlined in the Singapore Green Plan 2030, we aim to seize opportunities and contribute to the global transition towards renewable energy,” he says.
For 2025, the company has several big plans.
“Our plan is to start our Series B fundraising and scale up production,” Li closes.
—
Image Credit: Cosmos Innovation
The post As demand soars for alternative solar panel materials, Cosmos Innovation uses AI to make the search faster appeared first on e27.