Summary
Meng Yuan is a Senior Applied Scientist and Research Fellow with a decade of experience advancing information retrieval, embedding interpretability, and evaluation methodology. With a PhD in Information Retrieval from the University of Melbourne and recent industry appointment at Oracle, she builds end-to-end pipelines that combine rigorous statistical benchmarking, bootstrapping, and advanced indexing to optimize domain-specific search. Her current work pushes RAG and LLM training protocols forward through automated user-persona generation and systematic LLM-driven user simulations, bridging research and production needs. She has applied these methods to cross-modal retrieval, fake-news detection, and search-as-learning, and often surfaces interpretability insights from embedding spaces. Based in Melbourne, she blends academic rigor with practical engineering and a quiet passion for astronomy and open knowledge sharing that informs her collaborative, impact-driven approach.
10 years of coding experience
The University of Melbourne
English, Chinese, Japanese