Athena is an AI-powered application designed to revolutionize learning by automatically generating personalized educational courses. Users simply provide a high-level topic they wish to learn (e.g. "Quantum Physics", "Introduction to Python", "History of the Silk Road") and specify their learning preferences (such as desired level, duration and focus). Athena then leverages a sophisticated multi-agent AI system to research the topic, structure a comprehensive course syllabus, write explanatory content, and generate interactive learning materials like quizzes, all tailored to the individual user's needs.
Athena - Where Knowledge is Crafted for You.
Traditional learning resources, whether textbooks, online courses, or tutorials, often adopt a "one-size-fits-all" approach. This fails to cater to the diverse needs of learners who have different backgrounds, goals, learning speeds, and preferences for how information is presented. Key challenges include:
- Lack of Personalization: Standard materials don't adapt to individual learning paces (too short or too long), desired focus (broad vs. deep dive), or preferred level (beginner vs. advanced).
- Information Overload & Structure: Learners often struggle to find reliable resources and structure them into a logical learning path, especially for complex topics.
- Passive Learning: Many resources lack integrated active learning components like quizzes or exercises, which are crucial for knowledge retention and understanding.
- Time Consuming: Manually curating a personalized learning plan and materials takes significant time and effort.
Athena addresses these challenges by employing a collaborative multi-agent system built using the AutoGen framework. This system intelligently automates the course creation process:
- Understanding Needs: Athena takes the user's topic and learning preferences as input.
- Intelligent Planning & Research: A
Course Planner Agentdesigns a logical course structure (modules, lessons). AModule Research Agent, equipped with web-search capabilities, gathers relevant and up-to-date information for each section. - Tailored Content Generation: A
Lesson Writer Agentsynthesizes the researched information and crafts explanations, examples, and analogies specifically tailored to the user's stated preferences (level, duration, focus). - Active Learning Integration: An
Quiz Master Agentgenerates quizzes of multi-choice or true/false format and other kinds of active learning activities based on the lesson content to reinforce learning. - Quality Assurance: A
Course Assemblerchecks the generated content for accuracy, appropriateness and user preferences, ensuring clarity, engagement, and relevance. It assembles these components into a cohesive, structured, and personalized course ready for the user.
By automating and personalizing course creation, Athena provides a dynamic, efficient, and effective learning experience tailored to the individual.
- NextJS with TypeScript and React 19
- shadcn/ui with TailwindCSS
- Azure functions for agentic workflow
- Azure Postgres
- Azure AI foundry
- Framework: Microsoft AutoGen
- Models: OpenAI o3 mini, gpt 4o mini
- Vercel for hosting the website
- Azure Function Apps
- Personalized Course Generation: Creates courses based on any user-specified topic and learning preferences (level, duration, focus).
- Dynamic Content Creation: AI agents write unique explanations, examples, and analogies.
- Active Learning Integration: Automatically generates quizzes and other exercises related to the content.
- Multi-Agent Architecture: Utilizes AutoGen to orchestrate specialized AI agents (Planner, Researcher, Writer, QuizMaster, Assembler).
- Built-in Quality Control: Includes agents dedicated to validating factual accuracy and critiquing pedagogical quality.
- Adaptable Knowledge Base: Leverages LLMs and web search tools to cover a vast range of topics.
- MultiModal course content (Images, Videos, Charts, etc.)
- Save/Enroll for courses
- User Progress for courses modules, lessons & quizzes
- Add Snippets for modules
- Custom Rendering Engine (for supporting different types of content like images, video, etc.)