Ryan Gu Email & Phone Number

Bellevue, Wa, Us

Berkeley, Ca, Us

• Developed LLM agent system using Autogen to generate sample customer data, analyze the data, and predict thesuccess of a product in a new market. Integrated open-source Generative AI models from HuggingFace to enhanceagent capabilities.• Used Parameter Efficient Fine Tuning methods such as Low Rank Adaptation (LoRA) to fine tune LLM agents in PyTorch, making generated data human-like and be able to conduct accurate sentiment analysis.

Portland, Oregon, United States

• Trained models for Retrieval Augmented Generation (RAG) to generate customized website code with up to 80% accuracy to save development time.• Cleaned, prepared, and vectorized customer data using HuggingFace embedding models to make data readable to LLM.

Part of the Energy Efficiency team in the UW SEAL Lab. Help companies optimize energy use to save money.