How Do Data Scientists Build Successful AI Models and Transform Industry?
The first half of the talk examines how Data Science teams operate within major global Consumer Packaged Goods (CPG) corporations. It will provide an overview of the problem space tackled by Data Science in the CPG sector, how value is measured, the techniques employed, and the essential skills required of team members. Additional topics include considerations for the technical stack, collaboration with internal and external teams, and the typical lifecycle of projects—from initial conception and proof of concept (PoC) to real-time, scalable implementations. The discussion will conclude with insights into how recent advancements, particularly in Large Language Models (LLMs) and Generative AI, are reshaping the field.
The second half of the talk focuses on how data scientists build successful AI/ML models, their priorities, and the structure of AI-driven projects. We will showcase two business use cases—Optimal Shelf Assortment and Text Mining for information extraction and generation from tables and images—to illustrate the practical application of AI in industry. Finally, the talk will offer a perspective on how industry collaborations with leading universities can transform cutting-edge scientific algorithms into integrated business solutions that deliver tangible value.
