Yutong Wu

• Developed a fully local, privacy-preserving Retrieval-Augmented Generation (RAG) chatbot to query technical PDFs containing both text and • Built a semantic search pipeline using sentence-transformers, ChromaDB, LlamaIndex and CrossEncoder for hybrid retrieval and reranking. • Designed and deployed an interactive Chainlit UI with document collection selection, 5-star rating, and user feedback. • Currently implementing a reward model to train the LLM Model .

• Developed an intelligent chatbot using Large Language Models (LLMs), specifically ChatGPT technology, to enhance customer interaction and improve efficiency in customer service. • Integrated LangChain for enhanced natural language processing and understanding. • Applied pattern recognition techniques to analyze customer interaction data

• Worked as a tutoring in the course: Algorithm of Numerical Linear Algebra • Supported in the coding course and helped students with their problems in algorithm • Helped with the homework during the semester and every two weeks provided students with the corrections of their homework with comments

• Worked in a group leading by a professor with the direction of 5G techniques • Discovered the basic concepts in cloud computing, and also some developments like fog computing in both models and formulas • Presented a brief conclusion every week

• Ensured that every trade of the supermarket went well and dealt with over 50 trades every day • Worked with international students and workers to support the entire operation of the store • Assimilated into the local environment meanwhile shared different cultures with people around the world and built networks

Seminar : Histology and Deep Learning Schwerpunkte : Deep Learning, Medical Image Processing, Pattern Recognition

Noten : 86/100 Abschlussarbeit : Basic Image Processing GUI Based on Matlab Schwerpunkte : IoT, Cloud Computing, Communication Engineering