Highlights
- A policy-gradient deep RL agent selects each learner's next challenge.
- A GRU / BiLSTM models learning style, debugging habits and pace.
- A GNN similarity graph transfers effective learning paths between learners with different backgrounds.
- A RAG task generator turns anonymized real mistakes into debugging exercises; an AI reasoning panel explains every recommendation.
Built with
Deep RLGRU / BiLSTMGNNRAGPythonFlaskNode.jsHTMLCSS
Next project
PathLearnCareer-aligned AI tutor