Textbook
A direct link to the current textbook, with PDF download when that file is configured.
CS321M
The course hub for AI Measurement Science.
Everything for CS321M in one place.
Course materials
Slides, readings, and syllabus, delivered to your inbox.
Sign up to get the full CS321M materials library. Materials are posted as the quarter progresses.
Slides
A growing lecture deck archive as materials are posted through the quarter.
Syllabus
The live course arc, readings, events, and learning outcomes in one place.
Course arc
Four modules from theory to practice.
CS321M moves from measurement foundations through statistical methods to evaluation design and open infrastructure.
Module 01
Constructs and proxies
Defining what you're measuring and how benchmarks fall short.
Module 02
Uncertainty and statistics
Variance, confidence, and sampling built into every evaluation.
Module 03
Evaluation design
Protocols for fair comparison and reproducible results.
Module 04
Open resources
Textbooks, code, competitions, and public materials.
Beyond the classroom
The Predictive Evaluation Competition
The competition is the empirical testing ground for ideas developed in CS321M and the broader AIMS community. Teams build sample-efficient measurement systems and test them against shared benchmarks.
Stanford course listing
Visit Stanford CS321MFollow the work.
New research, course updates, and software releases. One email when it matters.