Inspiration
Many schools in the Democratic Republic of the Congo still manage learning, attendance, grades, communication, and payments using disconnected or manual systems. KELASI was created to provide one accessible digital platform adapted to the Congolese educational context.
What it does
KELASI connects students, teachers, parents, and school administrators.
Students can access lessons, exercises, results, announcements, and personalized learning support. Teachers can manage classes, grades, attendance, educational content, and identify students who need additional support. Parents can monitor their children's academic progress and receive practical recommendations. School administrators can follow institutional performance, attendance, payments, and important alerts.
KELASI AI analyzes educational information and provides simple, actionable guidance for students, teachers, parents, and school leaders.
How we built it
KELASI was developed with Flutter and Dart for a multiplatform experience. Firebase Authentication, Cloud Firestore, Cloud Functions, and Firebase services provide authentication, data storage, notifications, and secure backend operations.
OpenAI models power the educational assistance features. API requests are routed through secure Firebase Cloud Functions so that private API credentials are never stored inside the mobile application.
Codex and GPT-5.6 supported the Build Week development process by helping review the repository, debug data mapping and course retrieval issues, improve the architecture, expand technical and vocational learning support, strengthen security, and prepare the project for testing and submission.
Challenges we faced
The main challenges included organizing information for several user profiles, aligning learning content with the Congolese curriculum, correcting data-mapping problems, retrieving the correct courses for each class, and protecting the OpenAI API credentials.
We also had to transform an existing educational application into a clearer, more secure, and demonstrable Build Week project.
Accomplishments
We created a working educational platform adapted to the DRC, with dedicated experiences for students, teachers, parents, and school administrators.
We also implemented secure AI integration, personalized educational analysis, school-management features, technical and vocational learning support, and a documented GitHub repository.
What we learned
We learned how AI can support an entire educational community when its recommendations are contextual, understandable, and actionable. We also improved our approach to application security, data architecture, testing, and AI-assisted development with Codex.
What's next for KELASI
Our next steps are to test KELASI with more Congolese schools, expand curriculum content, improve offline access for areas with limited connectivity, add more local-language support, and build stronger analytics for teachers and school administrators.
Log in or sign up for Devpost to join the conversation.