Introducing lolpop: an Open Source Framework for Machine Learning Workflows

Machine learning teams have produced some excellent tools in the last decade to assist data scientists and machine learning engineers in their day to day work. However, teams are often left to their own devices when putting all the blocks together to build workflows. This introduces a lot of complexity, leaves workflows prone to error, and exacerbates tensions between teams due to fundamental differences in how they operate. lolpop (https://github.com/jordanvolz/lolpop/) is a new open source software engineering framework I built for machine learning workflows. The overarching goal is to provide a framework that can help unify data science and machine learning engineering teams. We believe by establishing a standard framework for machine learning work that teams can collaborate more cleanly and be more productive. In this talk, we'll introduce the tool, the motivations behind it, and demonstrate how it can begin to help teams streamline their ML use across all their use cases.