This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle
What We Built
We built AI Teacher Assistant, a web application designed to help teachers, tutors, and educators prepare educational materials more efficiently.
Teachers spend a significant amount of time preparing lecture notes, creating assignments, generating questions, and organizing classroom resources. These repetitive tasks can take time away from teaching and interacting with students.
Our project aims to simplify this process by bringing educational content generation, document management, editing, and PDF export together in one platform.
Demo
Code
GitHub Repository: https://github.com/AI-Teacher-Assistant/AI_TEACHER_ASSISTANT
This project combines backend development, database integration, authentication, educational content generation, document management, and PDF generation into one application.
How We Built It
We developed AI Teacher Assistant using:
- Node.js and Express.js for the backend and API routes.
- EJS, HTML, CSS, and JavaScript for the frontend and interactive user interface.
- MongoDB and Mongoose for database operations and data persistence.
- Google Gemini API for AI-powered educational content generation.
- A built-in educational content generator that provides a fallback when the Gemini API is unavailable.
- JWT and HTTP-only cookies for authentication.
- bcrypt for password hashing.
- PDFKit for generating downloadable educational PDFs.
- Automated integration and edge-case tests to validate important application workflows.
We designed the application so educators can move through the complete workflow: generate educational content, review and edit it, save it, and export it for classroom use.
The built-in fallback generator also allows the application to generate structured educational materials without requiring a Gemini API key, although its output is template-based rather than equivalent to a full language model.
Why Does Open Innovation Matter?
Open innovation creates opportunities to build AI-powered tools that are more flexible, customizable, and accessible.
For educational applications, open-weight models may provide greater control over deployment, enable experimentation with specialized educational tasks, and potentially reduce inference costs depending on the model and hosting setup.
Our next step is to evaluate and integrate a suitable open-weight model into AI Teacher Assistant, compare its output quality with our current implementation, and explore how it could make educational content generation more affordable and adaptable.
We believe combining practical educational software with open AI technologies can help us create useful tools for educators while giving us more control over how AI is integrated into our application.
Team Submission
This is a collaborative project built by a team of four members.
Team Members:
- @prachimanchandapm_20_c76a
- @sunaina_2d2daf53cc2e7c35d
- @mahak_df3f16062dee3d361da
- @shivalika_fc425e435190b9
We worked together to bring the frontend, backend, authentication, AI integration, document management, and PDF generation into one educational application.
Top comments (1)
This project is another invention from ideas. This project is indeed very special for me.