Xingtai Li

Built generative AI models (Firefly) and LLM-based assistants, optimizing diffusion training/inference to scale to 7B+ creations and delivering secure, production-ready generative workflows used by millions.

Built Nordstrom's core ML platform to automate training, deployment, and monitoring, reduced deployment time from weeks to minutes, and delivered real-time pipelines powering recommendations and fraud detection for millions of users.

Built large-scale Spark/Kafka ML pipelines for insurance pricing, fraud, and claims prediction, implemented real-time risk scoring and CV models that reduced claim cycle time by 40% and automated inspections.

Completed a master's program combining computer science and statistics with coursework and projects focused on machine learning and data analysis.

Earned a bachelor's degree in Biological System Engineering with coursework in engineering principles and quantitative analysis applied to biological systems.