Ross Katz

Ross Katz is a Principal & Lead Data Scientist at CorrDyn, a data consultancy, where he builds and oversees the development of production machine learning and NLP systems that solve real operational problems. He led the team that designed and built the fine-tuning, evaluation, and deployment pipeline described in this talk, which runs in production today processing thousands of documents daily. Ross holds a Master of Information and Data Science from UC Berkeley and has spent almost 20 years working at the intersection of data science, strategy, and analytics across edtech, e-commerce, SaaS, marketing, and non-profit sectors.

Presenting

• Your Data Is Too Sensitive for an API: Fine-Tuning an Open-Source LLM for Production Document Parsing