10th K AI: I Built an AI Tutor for My Sister Who Was Stuck on Textbook Questions

For the Hacktoberfest Weekend Challenge, a young coder created an AI tutor to assist his sister, a Class 10 student, with textbook questions she found difficult to understand. He noticed that even when the information was in the textbook, students struggled because the language was confusing or the questions were phrased in ways that were hard to grasp.

To address this, he built "10th K AI", an AI-powered Karnataka SSLC Class 10 textbook tutor that not only provides answers but also helps students understand the concept, identify the question pattern, and offers solutions and memory tricks. The AI retrieves relevant textbook passages, interprets them into a structured learning format, and generates question-specific examples for the student to practice.

The AI system utilizes Retrieval-Augmented Generation (RAG), which combines keyword search, semantic search, and a locally run open-source LLM (language model) using the Ollama platform. This setup ensures the tutor can be customized and run independently from proprietary AI APIs.

10th K AI is designed for students to learn patterns in question types, such as recognizing person-contribution relationships, cause-effect patterns, and formulas that need calculation. The goal is to empower students to quickly identify question types and apply the appropriate method, rather than simply giving the answer.

Initially developed for his sister, the tutor eventually became a tool that could help any student struggling with textbook questions. The project demonstrates that a focused, specific problem can lead to meaningful technological solutions, rather than pursuing generic AI concepts. The coder learned that solving real-world challenges can be far more impactful than merely developing an intriguing AI idea.