-
SmartEdu AI Tutor
Inspiration
As a student myself, I identified that engineering students often struggle to find immediate, step-by-step guidance for complex technical topics. I wanted to build a tool that bridges this gap, allowing students to learn effectively and "push the envelope" in their academic journeys.
What it does
SmartEdu AI Tutor is a multimodal web application designed to provide instant, personalized tutoring. It allows users to upload images of their doubts, which the AI analyzes to provide clear, step-by-step explanations in real-time.
How we built it
Frontend & Interface: Built using Streamlit for a fast, responsive user interface.
Language: Developed entirely in Python.
AI Engine: Powered by Google Gemini 3.5 Flash for intelligent, fast Q&A.
Multimodality: Integrated Pillow (PIL) for efficient image processing.
Deployment: Hosted on Streamlit Community Cloud for easy access.
Challenges we ran into
The primary challenge was configuring the deployment environment. Managing dependencies and ensuring the application could securely access the Gemini API required troubleshooting, which was a great learning experience.
Accomplishments that we're proud of
I am proud of successfully integrating multimodal image processing and deploying the application live, creating a tool that can genuinely assist fellow engineering students.
What we learned
I learned the importance of environment configuration in cloud deployment, how to effectively manage project dependencies using requirements.txt, and gained practical experience in integrating LLMs into a real-world, multimodal application.
What's next for SmartEdu AI Tutor
Future plans include adding support for more complex file types, incorporating voice-based interaction to further personalize the tutoring experience, and optimizing the AI's response time for even faster feedback.
Log in or sign up for Devpost to join the conversation.