SujhBujh

Inspiration

We were inspired by the everyday societal problems that often go unnoticed or remain unresolved because there is no proper connection between the people facing them and the organisations capable of solving them. We realised that universities have research capabilities, students have innovative ideas, and industries have the resources to implement solutions, but these stakeholders often work independently. We wanted to create a platform that brings them together and turns real-world problems into opportunities for collaborative innovation.

What it does

Sujh Bujh is an AI-powered societal innovation collaboration portal focused on Jharkhand. It allows citizens to report local challenges through a simple platform. The system uses AI to analyse and categorise complaints, identify potentially duplicate reports, and match problems with relevant institutions based on their expertise. It also helps surface existing research and possible solutions, reducing repeated efforts. A centralised dashboard provides an overview of reported challenges and their status.

How we built it

We built Sujh Bujh using React.js, Vite and Tailwind CSS for the frontend, and Node.js and Express.js for the backend. We integrated the NVIDIA API with the GPT-OSS-20B model to power AI-based complaint analysis and classification. Separate backend services handle complaint processing, duplicate detection and institutional routing. We used structured JSON datasets to maintain complaint records, institutional information and solution references. The frontend and backend were deployed on Render to make our prototype accessible online.

Challenges we ran into

One of our biggest challenges was converting unstructured, real-world problem descriptions into meaningful and actionable information. Designing AI prompts that could consistently classify complaints and return structured responses required experimentation. We also faced challenges in identifying duplicate complaints, matching problems with suitable institutions and connecting different frontend and backend components. Ensuring that the AI-assisted workflow remained understandable and useful rather than simply producing automated outputs was another important consideration.

Accomplishments that we're proud of

We are proud of building a working prototype that brings together citizen problem reporting, AI-powered complaint analysis, duplicate detection and institutional routing in one platform. We successfully integrated an external AI API into our backend and developed a dashboard to provide visibility into reported problems. Most importantly, we transformed our initial idea into a functional, deployed application that demonstrates how technology can connect societal challenges with potential solutions.

What we learned

Throughout this project, we gained practical experience in full-stack development, API integration, AI prompting, backend service design and deployment. We learned how to work with AI-generated outputs, structure unorganised information and connect multiple components into a single application. We also understood the importance of designing technology around actual user needs and the value of collaboration, communication and dividing responsibilities within a team.

What's next for Sujh Bujh

We aim to expand Sujh Bujh into a more comprehensive civic innovation ecosystem. Our future plans include multilingual voice-based reporting, image-based problem identification, geolocation-based complaint mapping, impact-based prioritisation and real-time notifications. We also envision dedicated collaboration spaces where universities, students, researchers and industry partners can work together on assigned challenges. Integrating a scalable database, institutional authentication and progress tracking will help us move towards a platform that can support real-world implementation across Jharkhand and eventually other regions.