Hugo Bowne-Anderson
Data Science Educator and Vanishing Gradients Host
Hugo Bowne-Anderson is an independent data and AI consultant with extensive experience in the tech industry. He is the host of Vanishing Gradients and High Signal podcasts, exploring developments in data science and AI. Previously, Hugo served as Head of Developer Relations at Outerbounds and held roles at Coiled and DataCamp, where his work in data science education reached over 3 million learners. He has taught at Yale University, Cold Spring Harbor Laboratory, and conferences like SciPy and PyCon, and is a passionate advocate for democratizing data skills and open-source tools. He also regularly teaches courses on Building LLM Applications for Data Scientists and Software Engineers.
May 13th, 12:05 – 12:35PM ET
Evaluation-Driven Development: Building Reliable AI Systems
Too many AI projects stall out in proof-of-concept purgatory — impressive demos that never become reliable products. In this session, we’ll explore how to escape that trap through Evaluation-Driven Development: an approach that puts evaluation at the core of building AI and LLM applications. You’ll learn how to bootstrap evaluation frameworks even before real users arrive — from generating synthetic user queries and manually labeling outputs to using LLMs as judges, building evaluation harnesses, and introducing lightweight observability to track accuracy, cost, and latency. Evaluation-Driven Development isn’t an optional add-on — it’s the foundation for shipping AI systems that work.