a lightweight, AI-assisted web application designed to help students manage academic stress by turning unstructured academic text into clear, editable tasks and a realistic study plan.
Inspiration: This project was inspired by my own experience feeling overwhelmed when starting assignments or projects. Even when I know what needs to be done, the mental load of organizing tasks and figuring out where to start can be stressful and discouraging. I wanted to create a tool that helps reduce that initial anxiety by breaking large, unstructured text into clear, manageable steps, while using supportive and non-judgmental language.
What it does: Academic Overwhelm Assistant helps students turn overwhelming academic text into clear, manageable steps. Users can paste in a syllabus, assignment description, or project outline, and the app uses AI to extract tasks and generate a realistic plan. Tasks can be edited, and the plan adapts when the user marks that they are behind. Throughout the experience, the app uses supportive, anxiety-friendly language to reduce stress and encourage progress without pressure.
How we built it: The application is built using Python with FastAPI for the backend and Jinja2 + HTMX for a lightweight, responsive frontend. The user pastes in academic text such as a syllabus or assignment description, which is sent to the Gemini AI API to extract tasks and generate a realistic plan. Tasks can be edited, and the plan updates dynamically. An “I’m behind” feature allows the schedule to adjust while displaying reassuring, anxiety-friendly messaging.
Challenges we ran into: One of the main challenges was handling AI API setup, version compatibility, and quota limitations while ensuring the application remained stable. Debugging environment and dependency conflicts also required careful troubleshooting. Designing the project to feel calm and supportive, rather than overwhelming was another challenge, especially when balancing functionality with simplicity under a tight time constraint.
Accomplishments that I'm proud of:
What I learned: through this project, I learned how to integrate an AI model into a full-stack application and design interactions around user well-being rather than productivity pressure. I gained experience working with FastAPI, template rendering, and asynchronous workflows, as well as handling API configuration, quotas, and error states. I also learned the importance of designing software that adapts to users when they fall behind instead of punishing them.
What's next for Academic Overwhelm Assistant: In the future, this project could be expanded with calendar integration, longer-term progress tracking, and more personalized planning based on a user’s workload and preferences. Additional accessibility features and customization options for language tone could further support users with different needs. With more time, the AI could also learn from user feedback to improve task breakdown accuracy and scheduling suggestions.
Log in or sign up for Devpost to join the conversation.