AI Engineering Reading Notes — Chapter 3: Evaluation Methodologies
How do we know an AI application is good enough? Notes on measuring predictions, testing outcomes, comparing answers, and evaluating the judges themselves.
How do we know an AI application is good enough? Notes on measuring predictions, testing outcomes, comparing answers, and evaluating the judges themselves.
Understanding what shapes a foundation model: the data it learns from, its architecture, how it is trained to answer, and how it selects output tokens.
Reading Chapter 1 of AI Engineering: how language models learn, what makes a foundation model useful, and why adapting an existing model is a practical starting point.