Open Category Submission

832 – Brains, Bots, and Builds: Generative AI for Hands-On STEM Learning Through 3D Printing

This project introduces an AI-powered educational platform that will be combination of generative AI and 3D printing to deliver gamified and personalized engineering challenges for junior high school students. Students are often disengaged in formal STEM education and the level of difficulties seems high with the lack of active engagement. This project aims to create adaptive, real-world tasks tailored to each learner’s skill level, learning pace, and creative potential. The main of the proposed system is the generative AI algorithms which can generate engineering design challenges and also provide real-time feedback on 3D models to offer personalized learning paths. Students are aimed to by receive a unique challenge (e.g., designing a flood-resistant bridge or a Mars rover component). The AI then generates a suggested 3D model and evaluates student designs to recommend modifications based on design integrity, material efficiency, and creative innovation.
The Gamification elements such as progress dashboards, interactive leaderboards, and rewards will be used for active participation in classroom and maintain high levels of engagement. The system also encourages collaborative learning, using AI to match students with complementary skill sets for team-based design tasks. From a technical perspective, the platform combines generative modeling, machine learning-based design evaluation, and AI-enhanced analytics to support students and teachers. This aligns with ethical AI principles by ensuring transparency, inclusivity, and fairness in content generation and evaluation. The potential impact is that this can transform STEM learning into tangible and hands-on experiences to promote creativity and problem-solving. Moreover, this can reduce educator workload through automated assessments. The adaptive nature of the platform will ensure equal access and personalized pathways for making high-quality STEM education more inclusive and scalable. This project reimagines how we teach, learn, and create in STEM.

This project was inspired by the challenge I observed while teaching and mentoring students in STEM-related learning environments. I am teaching assistant at RED program which is training research skills for junior high school students in Tokyo through STEM education. Many students, especially at the junior high level, struggle to stay engaged with STEM concepts. While hands-on learning suppose to increase retention and motivation, most classrooms lack the capacity for personalized and project-based approaches. In RED program, we are starting to use 3D printer but have not used yet for personalized learning. I wanted to address this disconnect by blending generative AI, gamification, and hands-on 3D design into an accessible learning experience that would feel both exciting and meaningful to students. The underlying hypothesis is that combining generative AI with gamified 3D-printed engineering challenges will boost student engagement, deepen conceptual understanding, and improve learning outcomes. Generative AI offers a unique opportunity to provide personalized learning experiences at scale which traditional education struggle to achieve. By generating real-world engineering tasks, receiving feedback, and incorporating elements of choice, competition, and creativity, students are not only encouraged to solve problems but also to design and reflect, which are essential STEM skills.This direction is based on three core beliefs:

Students learn best when challenges are relevant and hands-on.

Gamified environments can significantly increase motivation, especially in younger learners.

Generative AI can democratize high-quality, tailored instruction and feedback.

By anchoring these beliefs in a structured learning framework with real-world applications, the project is designed to be scalable, inclusive, and impactful.

This project leverages a blend of generative AI, 3D modeling platforms, and gamified learning design to create a hands-on, personalized STEM learning experience for junior high school students. The system uses generative AI (e.g., similar models to PaddlePaddle) to dynamically create engineering challenges and give feedback to individual learners. These challenges are delivered through an interactive interface connected to free or low-cost 3D design tools such as Fusion 360, and linked to 3D printing workflows that bring student designs to life. To support development, key resources will include: access to open-source or API-based AI text/image generation tools, a platform for hosting interactive challenges and tracking user data, basic 3D printing hardware for prototyping and user testing in schools.
I plan to validate demand through pilot testing in schools firstly at RED program in Musashi Junior High School, gathering feedback from students, teachers, and parents. Early indicators of market need will be rising interest in maker education and the integration of AI into classroom and suggest strong alignment with emerging educational priorities. The core functionalities of this project will include: personalized AI-generated challenges with real-world relevance, design evaluation feedback powered by AI (e.g., structure, efficiency, creativity), gamified progress tracking with rewards, rankings, and missions, and collaborative mode, where students are paired based on complementary skills. To ensure a positive user experience, the interface will be simple, playful, and student-centered, with clear tutorials and adaptive difficulty levels. Success metrics will include engagement rate, challenge completion rate, quality of student output, and qualitative feedback from users and educators. Ultimately, the project is designed to be scalable and accessible, equipping students with both STEM skills and AI literacy through meaningful creation.

This project enhances STEM education by combining generative AI, 3D printing, and gamification to create a dynamic, personalized learning environment. The innovation lies not just in the technologies used, but in how students are integrated to support adaptive learning and creativity. Unlike traditional STEM approaches that often rely on static content and one-size-fits-all challenges, this solution uses generative AI to create custom engineering tasks based on individual learner profiles. Each challenge is framed as a real-world scenario such as designing disaster-resilient structures or space exploration tools and this method will encourage students to think like engineers, not just students. The AI also provides instant feedback on their designs, highlighting structural flaws or creative optimizations for continuous improvement. Creatively, the platform integrates gamified elements such as quests, digital badges, and leaderboards to keep students engaged and motivated. It also introduces AI-guided peer collaboration, enabling students to work together on projects by matching complementary skills and interests. This is rarely seen in current AI-education models. Innovation is further reflected in the low-barrier design by using accessible 3D modeling tools and affordable printing options. The system can be adopted by any schools with limited tech infrastructure. The combination of personalized AI guidance, tangible creation through 3D printing, and motivation through gamification is a forward-looking response to the disengagement and rigidity often found in STEM education. By fostering curiosity, creativity, and critical thinking, this project not only equips students with essential 21st-century skills but also reshapes how they see themselves as problem-solvers and future innovators.

This project is designed with scalability at its core by using modular, accessible technologies that can grow with demand. The platform will leverage cloud-based generative AI services and integrates with widely available 3D modeling tools (e.g., Fusion 360) to reduce development costs and infrastructure barriers. This ensures that schools with different levels of resources can even implement. To address potential bottlenecks in content generation or feedback delivery, the platform will use batch processing and AI caching techniques to pre-generate challenge templates and feedback options, minimizing real-time latency. To scale adoption across diverse educational environments, this project will develop teacher toolkits, onboarding modules, and curriculum-aligned content. These resources will empower educators to integrate the system into existing lessons while maintaining flexibility for future customizations. Sustainability is addressed through both environmental responsibility and user retention strategies. Environmentally, the platform promotes efficient 3D printing practices by encouraging students to optimize their designs for material use for reducing waste. Partnerships with schools can enable use of recycled filament or shared printers. To foster long-term engagement, the system incorporates progress tracking, personalized rewards, and collaborative features that grow with the student. As learners advance, the platform adapts to offer more complex, interdisciplinary challenges by keeping the experience fresh and developmentally aligned. Feedback loops with educators and students will continuously inform updates for ensuring the solution evolves alongside user needs and tech trends. In summary, this solution is designed not just to launch, but to grow responsibly, adapt intelligently, and engage learners deeply in a sustainable changing educational landscape.

This project is rooted in the belief that access to quality STEM education should be equitable and empowering for all learners. By combining generative AI, 3D printing, and gamification into a single learning platform, the solution addresses several pressing challenges in learning STEM. The platform can enhance the lives of its primary beneficiaries students by offering customized learning experiences that match their individual needs, interests, and strengths, regardless of their background or academic level. Through low-cost and open-source design tools, the system can be deployed in low-resource classrooms with minimal infrastructure and closing the digital divide and bringing innovative STEM learning to a broader audience. Inclusion is built into the core design. The AI-generated challenges incorporate culturally relevant content, and the platform provides options for accessibility features, and diverse character representations. Collaborative learning features are designed to promote teamwork across different ability levels and helping students build communication and empathy alongside technical skills.

Social impact will be measured using both quantitative and qualitative metrics, including: increases in student engagement and participation across diverse demographic groups, completion rates of AI-generated challenges, feedback from students, teachers, and parents on perceived confidence and inclusion and tracking access equity by geography, gender, and socioeconomic background. To ensure the solution remains responsive, this project will implement feedback channels and periodic review cycles with educators. These insights will inform challenge content, AI behavior, and platform updates to adapt for changing student needs and social contexts. By fostering curiosity, equity, and empowerment, this project not only supports individual learners but contributes meaningfully to broader goals of social justice and inclusive innovation in education.

Personal Information Collection Statement (PICS):
1. The personal data collected in this form will be used for activity-organizing, record keeping and reporting only. The collected personal data will be purged within 6 years after the event.
2. Please note that it is obligatory to provide the personal data required.
3. Your personal data collected will be kept by the LTTC and will not be transferred to outside parties.
4. You have the right to request access to and correction of information held by us about you. If you wish to access or correct your personal data, please contact our staff at lttc@eduhk.hk.
5. The University’s Privacy Policy Statement can be access at https://www.eduhk.hk/en/privacy-policy.

• I have read and agree to the competition rules and privacy policy.