Yuqiu Gan

WORK HISTORY

EDUCATION

ABOUT YUQIU GAN

I am a Data Scientist with 4 years industry experience and hands-on experience designing, training and deploying end-to-end AI/ML systems in production. My work spans Machine Learning, Deep Learning, NLP, LLMs, RAG systems, and agent-based AI architectures, with a strong focus on turning unstructured data into structured, actionable insights.At Daitrix (ElectrifAi), I’ve designed and shipped solutions used in production by enterprise clients, including:BERT-based transaction classifier: trained and deployed via FastAPI + Docker, improving accuracy on unseen vendors from <30% to ~40%.Production RAG system for vendor grouping: integrated web search APIs, legacy DB queries, and LLM reasoning; improved accuracy from ~60% to ~90%.Agentic RAG prototype for spend classification: built custom tools for similarity search and ML inference; showcased to clients and helped win a major RFP.Vector database infrastructure: designed schemas in MongoDB and leveraged a fine-tuned BERT encoder for domain-specific embeddings, reducing costs and improving semantic retrieval.Core SkillsLanguages & Data: Python, SQL, C++, R, PySpark, Pandas, MySQL, MongoDB, PostgreSQLML/AI Frameworks: PyTorch, Scikit-learn, XGBoost, Hugging Face Transformers, LangChain, LangSmithSystems & Infra: FastAPI, Flask, Docker, AWS, Apache Airflow, Spark, GitExpertise: NLP, RAG, AI Agents, ML Pipelines, Model Deployment, Vector Databases, Data Science