Nikita Kozodoi is an Applied Scientist at AWS, where he works with companies across industries to design and build custom generative AI solutions as part of the AWS Generative AI Innovation Center. His day-to-day work centers on a challenge many organizations now face: moving GenAI projects from promising pilots to production systems that are dependable, measurable, and actually create business value.

Nikita holds a PhD in Machine Learning from Humboldt University of Berlin, and his background in competitive machine learning runs deep: he has earned 18 medals in Kaggle competitions. That combination of academic rigor and hands-on competition experience shapes his approach to applied AI, from careful experimentation and validation to pragmatic engineering decisions under real-world constraints. He is particularly interested in the gap between what models can do in a demo and what it takes to make them work at scale.