Chip Huyen

a few challenges 1. accuracy requirements: if you fail 10% at each step, it becomes like 90% failures after a few steps very quickly 2. tool use challenge: extremely hard to get the model to get the right params, both because of the tool's API and bc of human's internal knowledge. like, if someone says "work hours", AI might cut off it at 9am - 5pm, but humans might give a wider range 3.

How does building with foundation models differ from traditional ML? This session, drawn from her just-released bestselling O'Reilly book, AI Engineering, will explore key shifts, including the increasing challenges of evaluating open-ended outputs and the growing integration of product and engineering.

As data use cases become more complex, data platforms evolve to meet their needs. Data workloads run in different environments (local and production), in different modes (batch and streaming), and across different hardware (CPU and GPU).

As data use cases become more complex, data platforms evolve to meet their needs. Data workloads run in different environments (local and production), in different modes (batch and streaming), and across different hardware (CPU and GPU).

As data use cases become more complex, data platforms evolve to meet their needs. Data workloads run in different environments (local and production), in different modes (batch and streaming), and across different hardware (CPU and GPU).

As data use cases become more complex, data platforms evolve to meet their needs. Data workloads run in different environments (local and production), in different modes (batch and streaming), and across different hardware (CPU and GPU).

Open-ended evaluation is hard, and the number of evaluation tools has exploded in response to this challenge. However, if tools could solve evaluation, evaluation would have been solved by now. While the right tools can make your life easier, this talk discusses why you should think twice before...

Fresh data beats stale data for machine learning applications. This talk discusses the value of fresh data as well as different types of architecture and challenges of online prediction.