Compute for Climate Fellowship 2025

2025 Cohort Announced – Meet the Winners!

We are excited to announce 23 groundbreaking startups selected for the 2025 Compute for Climate Fellowship! Chosen from over 370 applications across 61 countries, these startups will receive funding in form of AWS credits, expert mentorship, and access to advanced AI and cloud computing resources to develop innovative solutions for the climate crisis. Their projects tackle pressing challenges in clean energy, carbon removal, zero-waste manufacturing, sustainable agriculture, resilient infrastructure, and more. The winners of the 2025 cohort are:

80 Acres Farms

80 Acres Farms operates cutting-edge indoor vertical farms that grow fresh, pesticide-free produce using 100% renewable electricity and 95% less water. Combining advanced technology with deep analytics, the company delivers longer-lasting harvests while significantly reducing food waste and agricultural emissions.

As part of its work with the Compute for Climate Fellowship, 80 Acres Farms is building a machine learning-powered genetic computing platform capable of accelerating precision breeding and improving genetic crop optimization. Their goal is to develop high-performance, climate-resilient crops with improved yield, reducing input waste and boosting food system sustainability.

Visit the website

Aionics
Aionics is pioneering the use of AI in materials discovery for electrochemical systems in aviation, mobility and energy storage. Their platform integrates artificial intelligence, quantum simulations, and automated testing to design next-generation battery materials that are safer, more efficient, and more sustainable.

Through the 2025 Compute for Climate Fellowship, Aionics is developing advanced computational models to accelerate the discovery of non-flammable lithium-ion battery electrolytes. By applying machine learning–driven interatomic simulations in the cloud, they aim to improve the speed and scalability of battery innovation for a low-carbon future.

Visit the website

Amini

Amini is a deep tech startup providing AI-native infrastructure to address data and compute scarcity across the Global South. By transforming satellite and ground data into actionable environmental insights, Amini supports decision-making for climate adaptation, sustainable agriculture and resilient supply chains in vulnerable regions.

As part of the 2025 Compute for Climate Fellowship, Amini is developing a generative AI foundation model tailored to land-based climate intelligence in Barbados. The model will leverage multi-temporal, multispectral satellite data.

Visit the website

Anthrogen

Anthrogen is developing large protein foundation models to discover ā€œde novoā€ proteins with transformative applications across human health and climate. At the intersection of AI and synthetic biology, Anthrogen is pioneering a new approach to carbon removal by engineering enzymes and microbes that capture atmospheric COā‚‚ and convert it into valuable materials such as polymers and fuels.

Using advanced gene-editing and AI-enabled protein design, Anthrogen enhances microorganisms’ ability to absorb and store carbon, providing a scalable and cost-effective solution for greenhouse gas removal. Its technology paves the way for sustainable manufacturing while supporting climate goals.

As part of the Fellowship, Anthrogen is building a generative AI foundation model based on a discrete diffusion framework to simulate protein evolution at scale. These simulations will power the creation of entirely novel enzymes tailored for climate-positive applications in clean chemistry and carbon-to-value conversion.

Visit the website

Aura Aero

Aura Aero is a French aircraft manufacturer developing electric and hybrid aircrafts to decarbonize air transport. Committed to contributing to a 55% reduction in aviation emissions by 2030, the company uses fully digital, cloud-based design processes to deliver next-generation aircrafts without compromising performance or safety.

As part of the 2025 Compute for Climate Fellowship, Aura Aero is building a high-performance cloud computing platform to simulate aircraft aerodynamics, structural components, and energy systems in the cloud.

Visit the website

Avalo

Avalo is an AI-powered agricultural technology company pioneering ā€œrapid crop evolutionā€ to build a more efficient, resilient, and sustainable global food system. By combining evolutionary biology with machine learning, Avalo’s platform accelerates natural plant development—uncovering climate-resilient genetic traits without using GMOs.

As part of the 2025 Compute for Climate Fellowship, Avalo is developing an AI-driven genomic selection system that uses large-scale simulations to optimize crop yields and accelerate plant breeding for climate adaptation.

Visit the website

Barnwell Bio

Barnwell Bio is a biotech company improving animal health and biosecurity through monitoring and metagenomic sequencing, focusing on commercial farming. By analyzing animal byproducts, its biosurveillance platform enables early detection of pathogens, helping farms respond proactively to infectious disease outbreaks.

As part of the Fellowship, Barnwell Bio is developing MetaVet, an AI chatbot that integrates large language models and predictive analytics to support proactive animal health management.

Visit the website

BeyondMath

BeyondMath is developing generative AI models that simulate the physical world for live aerodynamic optimization and sustainable vehicle design. Its platform is built on the market’s largest AI model for real-time, industrial scale engineering.

As part of the Fellowship, BeyondMath is developing a generative AI foundation model for sustainable vehicle design, supporting more efficient, low-emission transportation technologies.

Visit the website

Climate X

Climate X is an AI-enabled platform that simulates localized climate events and adaptation strategies to help financial institutions build more resilient real estate portfolios. By translating physical climate risks—like floods, heatwaves, and storms—into financial metrics, Climate X enables smarter investment, lending, and sustainability decisions.

As part of the Fellowship, Climate X is enhancing Spectra, its flagship tool that uses high-resolution digital twins to assess and adapt to climate risks at the property level.

Visit the website

Climavision

Climavision is redefining global weather forecasting through a proprietary network of high-resolution radars and its Horizon AI forecasting suite. By closing observation gaps and using high-performance AI models, it delivers hyperlocal, real-time forecasts to support critical decisions across energy, agriculture, and emergency management.

As part of the Fellowship, Climavision is expanding its ClimaHD platform to deliver sub-kilometer weather predictions, leveraging high-performance computing to improve accuracy and climate resilience.

Visit the website

CounterCurrent

CounterCurrent is decarbonizing the global shipping industry with AI-enabled ocean forecasts, satellite data, and autonomous sensors that optimize maritime routes in real time. Its platform helps fleets reduce fuel use, transit times, and COā‚‚ emissions, making maritime logistics smarter and more sustainable.

Through the Fellowship, CounterCurrent is enhancing its Hydra and Odysseus tools, using generative ocean forecasting to cut fuel consumption by up to 25% without changing ship hardware.

Visit the website

Cusp AI
Cusp AI is a frontier AI company accelerating the discovery of innovative materials. CuspAI is using its AI-enabled materials ‘search engine’ to discover and validate next-generation materials up to ten times faster than traditional methods – accelerating powerful solutions related to climate change, the energy transition and the future of computing. As part of the Fellowship, CuspAI is using its custom, generative AI model to design metal-organic frameworks (MOFs) for large-scale decarbonization.

Visit the website

Eavor

Eavor delivers next-generation geothermal energy with its closed-loop Eavor-Loopā„¢ system, which provides clean and reliable baseload power without fracking. Its modular design enables scalable geothermal anywhere in the world, helping meet rising demand for secure, dispatchable renewables.

As part of the Fellowship, Eavor is building a modular, cloud-native multiphysics solver that uses AI to optimize drilling, system design, and autonomous geothermal power deployment.

Visit the website

FortyGuard

FortyGuard is the world’s first temperature AI platform for cities, helping them mitigate extreme heat and improve energy efficiency. It offers a real-time heat mapping with 2-meter precision and enables targeted cooling strategies that reduce energy use by up to 30%.

As part of the Fellowship, FortyGuard is developing a generative AI foundation model that integrates satellite imagery and IoT data to help cities cut energy consumption, protect public health, and improve climate resilience.

Visit the website

Fuse Energy

Fuse Energy is a renewable energy company, accelerating the transition to net zero. By developing its own wind and solar farms and leveraging AI-enabled trading, it delivers 100% renewable electricity to homes and businesses.

As part of the Fellowship, Fuse Energy is developing an AI-enabled energy trading and risk management (ETRM) system for real-time monitoring and automated trading of clean energy generation.

Visit the website

Matnex

Matnex is a deep tech startup accelerating the discovery and development of sustainable materials using AI and quantum mechanics. Their platform rapidly screens material candidates using AI models validated by quantum simulations, cutting the time and cost of development for clean technologies.

As part of the Fellowship, Matnex is developing SCOUT, a structure-based machine learning platform capable of screening over 10¹⁰ material compositions. The team is also developing a custom generative AI foundation model to accelerate the identification of materials optimized for sustainability.

Visit the website

Muir

Muir AI’s product intelligence platform helps businesses reduce emissions and optimize costs by combining AI, geospatial data, and supply chain modeling. It enables high-resolution emissions analysis in minutes, transforming how industries like pharmaceuticals and semiconductors manage sustainability across complex supply chains.Ā 

As part of the Fellowship, Muir AI is building Omni-BoM, a universal AI system that automates product emissions and cost analysis through advanced Bill of Materials processing.

Visit the website

OnDeck AI

OnDeck AI is developing advanced AI tools for ocean conservation and maritime monitoring on a global scale. Its system integrates vision-language models and retrieval-augmented generation (RAG) to interpret underwater imagery without labeled training data, automating up to 90% of visual monitoring tasks for scalable and affordable ocean observation.

As part of the Fellowship, OnDeck AI is building a motion-grounded video reasoning model on cloud infrastructure. This system will interpret unstructured marine video data in real time to support conservation efforts and enhance ocean monitoring worldwide.

Visit the website

Pretred

Pretred is a clean-tech company turning recycled tires into circular, eco-friendly industrial barriers. Their technology diverts large volumes of rubber waste from landfills and offers a durable, low-emission alternative to concrete barriers used in construction, utilities, airports, and municipal projects.

As part of the Fellowship, Pretred is developing SaferCrash SIM, a machine learning tool that predicts material performance and optimizes barrier design. This AI-driven system supports circular reuse of rubber while helping reduce road fatalities through safer and smarter infrastructure.

Visit the website

Rainstick

Rainstick uses AI and bioelectricity to enhance seed performance and support regenerative agriculture. By developing ā€œbioelectric seed recipesā€ that mimic the natural effects of lightning, Rainstick helps crops grow faster, stronger, and more sustainably—without synthetic inputs or genetic modification.Ā 

Rainstick is the first Indigenous-led company to join the Fellowship. As part of the program, it is advancing its AI platform to create precision seed recipes that enhance yields, improve soil health, and strengthen the resilience of native ecosystems.

Visit the website

SXD

SXD is reimagining product design with AI and computer vision to eliminate material waste by design. Its platform generates zero-waste patterns and layouts, reducing material use by up to 69%, carbon emissions by 80%, and production costs by 55%.

By turning offcuts and production scraps into design opportunities, SXD helps industries like fashion, automotive, and consumer goods to transition from linear to circular systems—lowering costs, minimizing waste, and accelerating sustainable production at scale.

As part of the Fellowship, SXD is developing To Zero by SXD AI, an algorithm that converts traditional product designs into zero-waste versions using machine learning, computer vision, and combinatorial mathematics.

Visit the website

Windfall Bio

Windfall Bio develops nature-based solutions that harness methane-eating microbes (mems) to capture and convert methane emissions from farms, landfills, and industrial sites into valuable agricultural products, such as organic fertilizer. This approach targets methane, one of the world’s most potent greenhouse gases, and it helps to reduce emissions while supporting sustainable agriculture.

As part of the Fellowship, Windfall is developing an AI system that uses metagenomic data analysis to predict optimal microbial environments for maximum methane conversion efficiency.

Visit the website

Xcimer Energy

Xcimer Energy is commercializing laser-driven inertial fusion, the only fusion method with experimental validation, to deliver affordable, reliable, zero-carbon energy. By advancing excimer laser technology, known for its high-energy pulse capabilities, Xcimer is building the world’s highest-energy laser system to make fusion a practical and scalable energy source.

As part of the Fellowship, Xcimer is developing Iris, a digital twin platform designed to optimize laser performance and accelerate the development and deployment of fusion energy systems.

Visit the website

What is the Compute for Climate Fellowship?

The Compute for Climate Fellowship is a global R&D funding program that empowers climate tech startups to leverage advanced cloud computing and AI in the fight against climate change. The Fellowship was founded in 2023 by the International Research Centre on Artificial Intelligence (IRCAI), under the auspices of UNESCO, and Amazon Web Services (AWS). Each Fellow company accepted into the program builds a groundbreaking proof-of-concept (PoC) climate solution that is funded by AWS, with technical and scientific advisors from AWS and IRCAI.

Program Features
and Benefits
Grants in the form of AWS credits to cover the cloud computing costs of the PoC (up to $200k per startup)
Three month PoC build period with personalized technical support and mentorship
Access to advanced computing services, like quantum computing, HPC, and generative AI tools
Expert guidance in advanced computing, AI, sustainability, and ethics from IRCAI and AWS mentors
Media and visibility via speaking opportunities and PR (on a case by case basis)
Showcase of results across United Nations and international bodies
Winners will ā€œthink bigā€ and demonstrate innovation, leverage advanced cloud computing, and show potential for significant global impact across at least one of the nine climate solution areas:

1. Clean Energy

2. Low-Carbon Transportation

3. Sustainable Agriculture and Food

4. Circular Economy and Industry

5. Sustainable Buildings

6. Greenhouse Gas Management

7. Carbon Removal

8. Environment and Climate Risk

9. Indigenous Solutions to the Climate Crisis

New Ninth Climate Solution:
Indigenous Solutions to the Climate Crisis

In addition to the eight solution areas above, the Compute for Climate Fellowship is also seeking proposals for tech-enabled climate solutions that are built by, with, and/or for Indigenous Communities. The Fellowship recognizes that Indigenous People are important leaders in addressing the climate crisis. These communities around the world are both widely impacted by climate change, and at the same time have traditional knowledge, innovations, and practices that offer powerful solutions to the climate crisis and biodiversity loss. The Fellowship is eager to support Indigenous innovators and indigenous solutions to the climate crisis and is putting out a call for proposals.

More background about the call here →.

Who Should
Apply?
Your company is privately held and was founded within the last 10 years.
You have a technical team capable
of building with advanced cloud computing.

Your proposal incorporates innovative use of advanced cloud computing services such as:

AI/ML and Generative AI

High-Performance Computing

Autonomous Computing

Spatial Computing

Simulation Technologies

Quantum Computing

Time series data analytics

2025 Fellowship Process and Timeline

March 4th- April 6th
Applications open

April – June
Application review and interviews with finalists

JuneĀ 
Winners announced

June – August
Match winners with experts. Plan and scope PoCs.

September – November
Build PoCs

2025 Fellowship Chairs

Lisbeth Kaufman

Lisbeth Kaufman

AWS Head of Climate Tech – Startups & Venture Capital

DAVOR ORLIČ

DAVOR ORLIČ

Chief Operation Officer, IRCAI

Fei Chen

Fei Chen

Director of Advanced Computing Solutions, AWS

Cody Rasmus

Cody Rasmus

Program Lead, AWS Startups

Aidan O'Sullivan

Aidan O'Sullivan

Associate Professor in Energy and AI at UCL

Press and Media

Frequently Asked Questions

Who can I contact with questions?
Please reach out to info@ircai.org with the subject ā€œIRCAI AWS Fellowshipā€

I applied to the Fellowship before but wasn’t selected. Am I eligible?
Yes, please submit a new application and you will be considered for the next program.

What happens if I’m not selected for the Fellowship?
Startups that aren’t selected can apply for $5k in AWS Credits. Eligibility criteria includes: 1) Have not previously received AWS Activate credits of equal or greater value, 2) Self-funded or pre-Series B (most recent funding round must be within the last 12 months, if applicable), 3) Have a fully-functioning company website, and 4) Founded in the past 10 years. All AWS Activate credits are in USD and subject to the AWS Promotional Credit Terms & Conditions.

Will AWS build the PoC with my team?
Selected startups will be responsible for building the PoC. IRCAI scientific mentors and AWS technical expert mentors will advise on the design of the PoC, but will not build it. The startups will own IP resulting from the PoC.

Who can apply?
Private climate tech companies formed within the last 10 years, from any country.

How long is the PoC build phase?
2-3 months of development time.

Who owns the IP?
Startups retain ownership of all IP resulting from the POC.

Supported By

CONTACT

International Research Centre
on Artificial Intelligence (IRCAI)
under the auspices of UNESCOĀ 

Jožef Stefan Institute
Jamova cesta 39
SI-1000 Ljubljana

info@ircai.org
ircai.org

FOLLOW US

The designations employed and the presentation of material throughout this website do not imply the expression of any opinion whatsoever on the part of UNESCO concerning the legal status of any country, territory, city or area of its authorities, or concerning the delimitation of its frontiers or boundaries.

PRIVACYĀ  POLICY