From ML Engineering to AI Engineering: Navigating the Shift to Foundation Models with Chip Huyen

​Chip Huyen is a writer and computer scientist currently at Voltron Data, working on GPU-native data processing and open data standards (Ibis, Apache Arrow, Substrait). Previously, she built machine learning tools at NVIDIA, Snorkel AI, and Netflix.

​In this fireside chat, Chip joins Hugo Bowne-Anderson to explore the unique challenges and opportunities in productionizing foundation models compared to traditional machine learning approaches. As AI systems become increasingly advanced and open-ended, ML engineers must adapt their strategies and techniques to ensure reliable, efficient, and scalable deployments.

​Key topics of discussion will include:

• ​From closed-ended to open-ended evaluation: Developing robust evaluation methodologies for foundation models, which can generate novel outputs and exhibit emergent behaviors;

• ​From feature engineering to context construction: Techniques for effectively prompting and guiding foundation models to perform desired tasks across diverse domains;

• ​Adapting to unstructured data: Strategies for processing and integrating the vast amounts of unstructured data required to train and operate foundation models;

• ​Infrastructure and tooling challenges: Scaling compute resources, optimizing workflows, and building reliable pipelines for foundation model deployment;

• ​The evolving role of the ML engineer: New skills, collaborations, and best practices for succeeding in the era of foundation models and AI-driven products.

​Join us for a conversation on the forefront of AI engineering, and discover strategies for navigating the shift to foundation models in your organization.