Meina Qu

Led end-to-end Retrieval-Augmented Generation (RAG) system for document-based Q&A using LangChain, enabling context-aware responses over internal documents. Built automated ingestion pipelines for PDFs, Markdown, and TXT, with text chunking, embedding generation, and vector indexing using Chroma and Pinecone. Implemented semantic search and conversation memory for multi-turn queries, and tuned chunk sizes and similarity for higher retrieval quality. Deployed on cloud infrastructure with Docker and orchestrated tasks via FastAPI and Celery. Enhanced ingestion and retrieval reliability with monitoring and modular pipelines.

Participated in Baidu entertainment projects (Boiling Point, Nebula). Built Python-based web scraping for data collection, authored Django-based backend management and PHP APIs for data handling. Designed scalable storage with DynamoDB and Object Storage, developed a robust multimedia file storage system, containerized services with Docker, and implemented real-time data processing via Kafka. Implemented ElasticSearch-driven video search and integrated with Baidu Cloud Storage, contributing to a short-video platform.

Led the development of a cloud-based precise delivery platform and QA system enabling students to seek academic support. Developed an online payment system using MySQL and Redis for fast, reliable transactions. Collaborated on backend services to ensure scalability and robustness.