With the rapid advancement of large language models (LLMs), AI agents are emerging across various domains. Based on this trend, we designed QueryMinds: SQL Training System with LLMs—an enhanced learning platform that builds on traditional course systems while offering key innovations:
1. AI-Powered SQL Practice: An intelligent AI instructor generates diverse SQL exercises and automatically evaluates student submissions.
2. Natural Language Information Retrieval: Instructors can use plain language to search for course-related content, and the system intelligently delivers accurate results.
1. AI-Powered SQL Practice:
Student users input keywords to trigger prompt-engineered queries, enabling the LLM to generate topic-specific SQL exercises along with explanations. Upon submission, student answers are scored using an LLM-based SQL equivalence evaluator, which compares logic and structure rather than syntax alone.
2. Natural Language Information Retrieval
The system enables both instructors and students to automatically retrieve and analyze their own data, providing personalized insights through natural language queries integrated with Django views and MySQL-backed evaluation logs.
