Vili Krainz

Vili is a seasoned data scientist with 14 years of experience in various domains (business consulting, banking, insurance, gaming, automotive, manufacturing). Specialized in data science on structured data which naturally arise from running a business.

He is passionate about both R&D and applying mathematical concepts to solve real-world problems and seeing the impact of his work firsthand. Enjoys solving complex problems, extracting hidden relationships from data, and designing a problem solution from the bare-bone idea all through to the solution implementation.

He established and led data science departments in several prominent Croatian companies operating on the global market. Vili currently works as a freelance data science consultant for international clients.

Data science on structured data – the truth about the data & the scientists

Data science on structured data still resembles most of the practical data science use cases. As companies grow their business, they start to gather large amounts of data. However, with rising amounts of data, the patterns and insights become less and less apparent, while optimization opportunities grow. Despite being present for longer and having less new recent breakthroughs in the field, my experience shows there’s still a lot to be said about traditional data science methods.

I have also noticed a lack of understanding, resources and courses that cover topics like problem formulation and sample preparation which are crucial for creating value with data science. The goal of the presentation is to go over certain common pitfalls, repeat some lessons learned and try to establish a framework to approach a structured data science problem on the very general level.