I built an AI learning platform that doesn't stop at generating a roadmap.
A roadmap tells you what to learn.
But learning also requires knowing:
What comes first? Where should I learn it? Can I ask questions? And have I actually mastered it?
That's the problem I wanted to explore with SkillGraph AI.
I built it as an adaptive learning platform that brings several AI systems together into one learning workflow.
π§ Learning Path Architect
Generates prerequisite-aware learning paths and represents them as an interactive knowledge graph, so learners can see how concepts connect instead of following a flat checklist.
π Librarian Agent
Finds and curates relevant learning resources for the topics in the path, connecting the roadmap to actual material the learner can use.
π¬ RAG Tutor
A retrieval-augmented tutor that can answer questions using contextual learning material rather than relying only on a general-purpose LLM response.
π Assessment Agent
Generates topic-specific quizzes and assessments so learners can test their understanding as they progress.
πΈοΈ Interactive Skill Graph
The learning path becomes an explorable visual graph where prerequisites and relationships between concepts are visible.
The architecture is built around multi-agent orchestration, with specialized components handling planning, retrieval, tutoring and assessment.
Don't just tell people what to learn. Help them learn it, find the right resources, answer their questions, and verify that they actually understand it.
SkillGraph AI: AI-powered adaptive learning platform.
I built an AI learning platform that doesn't stop at generating a roadmap.
A roadmap tells you what to ...