Jocelyn Beauchesne
About Jocelyn Beauchesne
Jocelyn Beauchesne is an Engineering Manager, Machine Learning at Hyperscience, with extensive experience in machine learning and data science.
Company
Jocelyn Beauchesne is currently employed at Hyperscience, located in the New York City Metropolitan Area. He serves as the Engineering Manager for Machine Learning, a position he has held since 2022. Hyperscience is known for its advanced machine learning solutions, notably those that streamline and automate office functions utilizing multi-modal LLMs.
Title
Jocelyn Beauchesne holds the title of Engineering Manager, Machine Learning at Hyperscience. His role involves leading engineering and machine learning initiatives to drive technological advancement within the company.
Education and Expertise
Jocelyn Beauchesne achieved a Master of Business Analytics from the Massachusetts Institute of Technology (MIT), specifically from the Operations Research Center. He also holds a Master of Science in Engineering and Applied Mathematics from École Polytechnique, as well as a Bachelor of Science in Engineering, Applied Mathematics and Computer Science from the same institution. Prior to this, he studied Physics, Computer Science, and Mathematics at Lycée Montaigne Bordeaux.
Professional Background
Jocelyn Beauchesne has an extensive background in machine learning and artificial intelligence. Before his current role at Hyperscience, he worked as a Machine Learning Engineer at Abnormal Security and Rapid7, and served as a Research Assistant at MIT Sloan School of Management. He has also worked internationally as a Data Scientist at BNP Paribas in Paris, and served as a Second Lieutenant in the Gendarmerie Nationale in Martinique.
Achievements
In his role at Hyperscience, Jocelyn Beauchesne directed research efforts for a multi-modal LLM focusing on zero-shot information extraction. He successfully reduced model training time by an average of 60% through parallelization and early stopping. Additionally, he decreased the annotation burden on customers by 75% without affecting performance, evidencing significant improvements in efficiency and customer satisfaction.