Bright Tutor

BrightTutor – Project Story Inspiration In classrooms today, students often fall behind silently. One small misunderstanding in a math step or explanation can cause them to lose the thread of an entire lesson. Teachers rarely see where the misunderstanding began they only see the final incorrect answer. With millions of middle- and high-school students learning at different speeds, true one-to-one support is nearly impossible. We wanted to fix that. BrightTutor was born from a simple idea: what if every student could receive personalized, adaptive guidance instantly?

What BrightTutor Does BrightTutor recreates the experience of working with a personal tutor. Students speak their reasoning out loud, and BrightTutor analyzes not just the answer, but how they think through each step. It identifies the exact point where understanding breaks and builds a personalized learning path to fix it. Teachers receive a clear dashboard that shows mastery levels, weak concepts, and student progress — insights they never had before.

How We Built It We designed BrightTutor to work end-to-end: A clean user interface built rapidly using Lovable

A backend that processes spoken reasoning

An adaptive engine that generates personalized practice sessions

A teacher dashboard with verified question management

A progress summary that highlights strengths and areas to improve

Everything was built within a few hours during the DevFest Linz AI Hackathon, focused on one goal: make adaptive learning effortless and accessible.

Challenges We Faced The hardest challenge was not technical — it was defining the real educational problem. We had to understand why students fall behind, how teachers diagnose learning gaps, and how to model adaptive reasoning in a simple and intuitive format. Time pressure forced us to simplify, prioritize, and focus on what truly mattered: diagnosing reasoning, adapting practice, and giving teachers visibility.

What We’re Proud Of In a single day, we built a working prototype that: Analyzes student reasoning

Generates tailored questions

Lets teachers verify AI-generated content

Tracks mastery across lectures

Gives students actionable feedback

BrightTutor feels like a real product, not just a demo.

What We Learned We discovered how powerful it is to design AI around student thinking, not student answers. We also learned that adaptive learning only works if teachers remain part of the loop with tools to verify, correct, and understand the AI’s output.

What’s Next We plan to expand BrightTutor to more subjects, develop richer analytics for teachers, and integrate with platforms like Moodle so schools can adopt it easily. Our long-term vision is simple: make personalized, one-to-one learning accessible to every student every day.

Built With

• elevenlabs
• fastapi
• gemini
• lovable
• postgresql
• python
• sqlalchemy