It’s incredible how evolution has allowed data scientists to adapt to such practices as committing entire Jupyter Notebooks to Git, ignoring version control altogether (or worse, using Google Drive!), and treating every script as a one-off experiment. In this humorous and insightful talk, we’ll share the trials and tribulations of a Machine Learning Engineer tasked with guiding a team of brilliant but chaotic data scientists towards the promised land of Software Engineering Best Practices.

From the horror of seeing code linted for the first time to the shock of realizing that unit tests can actually be useful, we’ll discuss the common pitfalls and how to navigate them with humor, patience, and a healthy dose of empathy. We’ll also share practical tips and tricks for fostering collaboration, improving code quality, and ultimately, building robust and maintainable machine learning systems.

Join us for a lighthearted yet informative journey through the world of data science and software engineering.