Talk 1: The Future of Diversity: An Open Letter From Depop
A talk with Odie Edo-Osagie
In this talk, we’ll explore how our approach to diversity within our work mirrors our approach to diversity within our team. We will be measuring the power of diversity in the data science industry, using our recommendation system as a parallel and example.
Talk 2: Ranking search results and recommendations at Depop
A talk with Ben Ajayi-Obe and Mitch Watts
In this talk, we’ll explore the unique challenges and opportunities faced by the search and ranking team at Depop, a dynamic and growing marketplace. As machine learning scientists, our goal is to deliver frequent, high-impact improvements to our models while ensuring their reliability. However, working in a cross-functional environment with multiple team members developing new features and models in parallel adds layers of complexity. How do we ensure that our solutions are not only impactful but also seamlessly deployed to experiments, maximising their value? We’ll share insights into our approach, the infrastructure that supports rapid iteration, and strategies for overcoming the challenges of concurrent development.
Technical Level: High-level/overview
Supported by
