📅 2025-08-16 — Session: Optimized AI Teaching Material and Query Engine
🕒 18:30–20:30
🏷️ Labels: Ai Teaching, Query Engine, Curriculum Design, Python, Education
📂 Project: Teaching
⭐ Priority: MEDIUM
Session Goal
The session aimed to enhance the development and optimization of AI teaching materials and query engines, focusing on both educational content and technical retrieval systems.
Key Activities
- Set up OpenAI API key in a Jupyter Notebook to ensure compatibility with LlamaIndex and OpenAI SDK.
- Developed a framework for evaluating educational responses, including a taxonomy and scoring rubric.
- Designed a JSON schema for annotating QA responses to aid book assembly.
- Created structured prompts for retriever analysis focusing on book scope and exercises.
- Refined a CS101 curriculum blueprint emphasizing dynamic programming and algorithm analysis.
- Outlined a strategy for transforming raw content into AI teaching materials.
- Provided guidelines for topic modeling with bilingual corpora, focusing on clustering and multilingual considerations.
- Developed a Python function to create a query engine with multilingual support.
- Enhanced query engine design for LLMs with hierarchical parsing and token limits.
- Planned integration for a query engine and node management system.
- Introduced a script for generating daily FAQs from notes stored in an SQLite database.
Achievements
- Successfully set up and tested the OpenAI API key integration in Jupyter.
- Completed an educational response evaluation framework.
- Finalized a CS101 curriculum blueprint.
- Developed a comprehensive integration plan for query engines and node management.
- Created a daily FAQ generator script for streamlined content management.
Pending Tasks
- Further testing and refinement of the query engine integration plan.
- Additional validation of the educational response framework in real-world scenarios.