Announcing the winners of the June Innovation Challenge

We had an extraordinary set of projects created during our June 2025 Innovation Challenge Hackathon. It’s inspiring to see what happens when our community comes together to solve for real world problems! 231 participants worked on 49 projects, showing the amazing capabilities they’ve developed through their own learning and coming together to collaborate as professional AI developers. The Americas Azure team created this program to foster a diverse talent pool in partnership with a range of organizations supporting groups who are underrepresented in technology. For this event we had participants from Microsoft’s Black Partner Growth Initiative, Código Facilito, DIO, GenSpark, HACC, NASA Space Apps Chicago and Mountain View, Project Blue Mountain, Propel2Excel, TechBridge, and Women in Cloud.

Hackers were judged based on the performance of their team’s solution, pushing the boundaries of innovation, leveraging the full capabilities of Azure, and for implementing Responsible AI. The winning projects solved for AI use cases that ranged from addressing the social impact of property inheritance issues to the technical challenges of working with unstructured data to managing wildfire risk.

• AI for heirs’ property: Many communities struggle with issues around heirs’ property and losing property because of overdue taxes, blight, and other challenges. In order to nurture opportunity in under-resourced communities, local governments and nonprofits work to identify potentially at-risk properties, offer programmatic services to address legal and other issues, and fund and deploy projects. For the hackathon, we looked for solutions to strengthen and speed up this complicated work.
• Handling unstructured data: As organizations experiment with techniques for augmenting Generative AI models, such as RAG, they can often see inconsistent results when trying to improve performance. A common issue is how they are chunking their data. We asked hackers to create a tool or process for handling unstructured data in a way that results in more consistent model performance.
• Wildfire resilience: Wildfires are growing in frequency and intensity, threatening ecosystems, infrastructure, and communities. These projects delivered solutions to help emergency responders, planners, and policymakers predict wildfire behavior, assess risks, and optimize response strategies—all in near real-time.

Every team who delivered a project has a lot to be proud of! We saw some great thinking, real imagination and overall a community with a strong growth mindset—engineers ready to solve big problems. It’s also worth noting this is our first Innovation Challenge Hackathon where hackers created multi-agent solutions. The pace of AI is fast! Here are the judges' picks for the winning prizes.

First place $10,000

InHeir.AI an intelligent and secure legal tech platform developed to streamline property dispute management and analysis legal professionals and local communities. The solution ensures data privacy, security and accuracy by adhering to responsible AI practices and responses are backed by established legal procedures.

Second place $5,000

Contextual Jigsaw Addresses a critical vulnerability in modern AI systems: how document segmentation for RAG (Retrieval-Augmented Generation) systems can systematically distort information, amplify biases, and mask malicious content. This project identifies and solves for the convergence of four interconnected threats that emerge when splitting documents into chunks, creating a perfect storm that compromises AI fairness, security, and reliability.

Heir Aid AI-powered platform designed to support legal professionals, nonprofits, and communities in resolving heirs’ property issues and protecting generational wealth. HeirAid streamlines the identification, analysis, and resolution of complex title and tax challenges, particularly in under-resourced neighborhoods.

Third place $2,500

Artemis Shield Wildfire Dashboard an AI-powered, cloud-native command and control platform designed to transform wildfire management from a reactive struggle into a proactive, data-driven strategy.

Ignis Sentinels a wildfire resilience platform powered by multi-agent AI system in near real-time. It helps emergency responders, planners, and policymakers predict fire behavior, assess threats to people and infrastructure, and coordinate smarter evacuations.

Ignis Map AI-powered wildfire prediction and response through real-time geospatial intelligence.

We look forward to our next Innovation Challenge hackathon later this year! There will be new tools and ideas to put to the test, and can’t wait to see what our community builds!