Inspiration
We believe learning shouldn't be a clash; it should be a conversation. We were inspired by the "I don't get it" moments students face when confronting dense, jargon-filled textbooks. We wanted to build a bridge—a tool that takes complex academic content and instantly translates it into accessible, relatable language. What it does
StudyBuddy ELI5 is a multimodal AI tutor. Users can upload a photo of a textbook page or paste complex text, and the app instantly provides an "Explain Like I'm Five" summary. It simplifies concepts using creative analogies—like explaining the laws of physics (F=ma) using the mechanics of a bicycle—to make learning intuitive and stress-free. How we built it
We built the app using Python and Streamlit to ensure a fast, clean interface. We integrated the Google Gemini API (gemini-1.5-flash) for its high-speed, multimodal capabilities. By using custom system prompts, we instructed the model to prioritize analogies and simplicity, ensuring it maintains scientific accuracy while demystifying difficult topics. Challenges we ran into
The primary challenge was balancing simplicity vs. precision. Initial iterations sometimes oversimplified concepts until they lost their meaning. We had to iterate on our prompt engineering to find the "sweet spot" where the AI remains academically rigorous while still being easy to digest for a beginner. Accomplishments that we're proud of
We are most proud of the seamless integration of image-to-text capabilities, which allows students to interact with physical textbooks directly. We also successfully implemented LaTeX rendering to ensure that mathematical equations and scientific notations remain clear and professional, no matter how complex the subject. What we learned
We learned the power of the Context Window. By effectively structuring how we pass information to the Gemini API, we learned to prevent the AI from "hallucinating" and kept the responses highly focused on the user's specific learning hurdle. We also gained hands-on experience in rapid UI development using Streamlit. What's next for StudyBuddy ELI5
We plan to add a "Progress Tracking" feature where the AI remembers which concepts a user has mastered, allowing it to build "knowledge paths" that become progressively more complex. We also aim to support audio-to-text, enabling students to record lectures and get summarized notes in real-time.
Log in or sign up for Devpost to join the conversation.