LlamaIndex is a developer-focused Python library designed to bridge the gap between large language models and private or domain-specific data. It offers multiple index types—such as vector, tree, and keyword indices—along with adapters for databases, file systems, and web APIs. The framework includes tools for slicing documents into nodes, embedding those nodes via popular embedding models, and performing smart retrieval to supply context to an LLM. With built-in caching, query schemas, and node management, LlamaIndex streamlines building retrieval-augmented generation, enabling highly accurate, context-rich responses in applications like chatbots, QA services, and analytics pipelines.

Agentic Shared Memory provides a robust solution for managing contextual data in AI-driven multi-agent environments. Leveraging vector embeddings and efficient data structures, it stores agent observations, decisions, and state transitions, enabling seamless context retrieval and update. Agents can query the shared memory to access past interactions or global knowledge, fostering coherent behavior and collaborative problem-solving. The library supports plug-and-play integration with popular AI frameworks like LangChain or custom agent orchestrators, offering customizable retention strategies, context windowing, and search functions. By abstracting memory management, developers can focus on agent logic while ensuring scalable, consistent memory handling across distributed or centralized deployments. This improves overall system performance, reduces redundant computations, and enhances agent intelligence over time.

Chat-With-CUHKSZ provides a streamlined pipeline for building a domain-specific chatbot over the CUHKSZ knowledge base. After cloning the repository, users configure their OpenAI API credentials and specify document sources, such as campus PDFs, website pages, and research papers. The tool uses LlamaIndex to preprocess and index documents, creating an efficient vectorized store. LangChain orchestrates the retrieval and prompts, delivering relevant answers in a conversational interface. The architecture supports adding custom documents, fine-tuning prompt strategies, and deploying via Streamlit or a Python server. It also integrates optional semantic search enhancements, supports logging queries for auditing, and can be extended to other universities with minimal configuration.