CraftPad: Enabling Human-AI Collaboration in Lesson Design
CraftPad is an AI-powered educational tool that help teachers generate lesson designs in a fast and high-quality way. It help teachers save time and create lesson design aligned with pedagogical theories.
Teachers spend a significant amount of time creating lesson plans, often struggling to maintain consistent quality across courses. CraftPad was designed as an AI-powered educational tool to streamline this process. It helps teachers generate structured, high-quality lesson designs aligned with pedagogical theories — saving time while supporting effective teaching practices.
Time Efficiency: Reduced lesson design time significantly.
Quality Improvement: Enhanced alignment with pedagogical frameworks.
User Satisfaction: Teachers reported a smoother and more reliable workflow.
Through research and teacher interviews, we identified three major challenges:
Time Intensive: Creating lesson designs consumes a large portion of teachers’ daily workload.
Inconsistent Quality: Quality varies depending on individual expertise, energy, and available resources.
Lack of Pedagogical Alignment: Lesson designs often lack systematic adherence to established teaching theories.
We explored ways to make the process more intuitive and flexible:
Refined Workflow: Teachers provide basic lesson details → AI generates → teachers review and refine → finalize and evaluate.
Whiteboard Format: A flexible, visual whiteboard interface that mirrors teachers’ natural editing and organizing habits.
Structured Input: Simplified input fields guide teachers unfamiliar with prompt engineering to provide the right context.
Modular Output: Lesson designs are broken down into sections (e.g., introduction, activities, assessment) so teachers can refine parts individually.
Teacher Control: Multiple ways to edit — section-by-section feedback, manual editing, or regenerating the entire plan — ensure teachers remain in control.
Edit Manually
Collaborate with AI
Ethical Safeguards with RAG: Integrated Retrieval-Augmented Generation (RAG) ensures outputs are grounded in a vetted knowledge base of pedagogical theories, minimizing bias and hallucination.
