DeepTutor

Overview

DeepTutor is a fully open-source agentic framework for personalized tutoring. It unifies citation-grounded problem tutoring with difficulty-calibrated question generation, using a hybrid personalization engine that couples static knowledge grounding with dynamic learner memory and extends to adaptive learning workflows, interactive books, and proactive multi-channel tutoring agents.

Key Features

• Citation-grounded problem tutoring and difficulty-calibrated question generation.
• Hybrid personalization engine coupling static knowledge grounding with dynamic learner memory.
• Continuous adaptation to each student's evolving needs.
• Extension of the personalization substrate to adaptive learning workflows, interactive books, and proactive multi-channel tutoring agents.
• TutorBench interactive benchmark with customized learner profiles grounded in university-level curricula across five domains.
• LLM-based first-person interactive evaluation protocol using a profile-driven student simulator.

Use Cases

• Personalized problem tutoring with citation-grounded responses.
• Difficulty-calibrated question generation for learner practice.
• Adaptive learning workflows and interactive books.
• Proactive multi-channel tutoring agents.

Before You Adopt

• License: Apache-2.0. Review its terms before using, modifying, or distributing the project.
• The project is presented as a technical report and work in progress, so adopters should treat it as evolving.