Inspiration
Teachers today face an overwhelming workload: lesson planning, student assessments, personalized guidance, and endless paperwork. As a result, they have less time to truly support individual students. We were inspired to build a tool that lightens this burden—using the power of AI that adapts to real classroom needs.
What It Does
TeachAI is an AI-powered platform that reduces teacher workload and enhances classroom efficiency. It offers:
- Lesson Planning Automatically generates lesson plans based on topics or curriculum standards, complete with clear learning objectives.
- Activity Designer Suggests engaging classroom activities aligned with the lesson and suitable for different grade levels.
- Test Builder Creates quizzes and exams automatically—customizable by topic, question type (multiple choice/open-ended), and grade level. Includes answer keys and scoring rubrics for each question.
- Student Assessment Tracks and visualizes student performance. AI gives personalized feedback and suggestions for each student.
- Teaching Methods Recommends various teaching strategies, step-by-step instructions, and pros/cons of each approach.
- Teaching Resources Suggests relevant teaching materials and references to support each lesson.
- Performance Analytics Visualizes student performance in graphs, with data export in CSV format for further analysis.
How We Built It
- Bolt was used to quickly build UI and LLM agents.
- Supabase stores lesson plans, tests, and student records securely.
- GEMINI 2.5 is integrated as our LLM engine for content generation and smart suggestions.
- Prompt Engineering ensures accurate responses aligned with educational context.
Challenges We Faced
- Crafting prompts that deliver accurate outputs based on grade level and subject.
- Managing LLM token limitations under the free usage plan.
- Designing an intuitive UI for teachers unfamiliar with technology.
- Integrating lesson planning, activities, and assessments into a seamless interface.
Accomplishments We're Proud Of
- Built a functional platform within Hackathon time constraints.
- Enabled teachers to plan lessons, create tests, and analyze students—all in a few clicks.
- Successfully connected data across lessons, tests, and student performance using AI agents.
What We Learned
- Gained deeper insight into the diverse responsibilities of real-life educators.
- Practiced UX design focused on accessibility and simplicity.
- Learned to manage structured educational data with Supabase and optimize LLM token usage.
- Experienced the synergy between AI, UX design, and education.
What’s Next for TeachAI
- Add image-upload or photo-based homework analysis by AI.
- Build an AI feature that recommends enrichment activities for individual students or small groups.
- Run pilot tests in real schools to collect feedback and iterate the platform further.
Built With
- bolt
- gemini
- supabase
Log in or sign up for Devpost to join the conversation.