Learn directly from AI Evals and Analytics

Good materials and knowledgeable instructors.

insightful

Class is well paced with working examples and instructors that make the content engaging and relatable.

This course offers a structured framework for developing critical thinking around AI evaluations. It provides practical guidance on aligning evaluation processes with product release cycles, identifying and tracking meaningful metrics, and defining clear pass/fail criteria. The curriculum also covers selecting appropriate evaluation techniques, determining when to involve subject matter experts and clarifying their roles, and making informed go/no-go decisions. In addition, the course addresses important areas that are often overlooked in other offerings, including experiment design, product monitoring, and an overview of open-source libraries for running evaluations. Stella and Amy were highly supportive, generously offering one-on-one office hours to discuss assignments and explore additional ideas. I sincerely appreciate their guidance and their willingness to invest time in supporting learners.

I genuinely felt empowered after completing this course. In the AI era, establishing an effective evals framework early, while keeping stakeholders aligned, is a critical step in building responsible AI products. AI may be a disruptive technology, but human subject matter experts still play a vital role in evals design and execution. Stella and Amy’s course strongly validated what I had believed for some time: that evals, risk awareness, and human judgment must remain central as AI products evolve. For a long time, I felt like I was navigating this new AI path alone. This course changed that. It brought together like-minded people who genuinely value, understand, and care about AI product risk, safety, and social impact. If you come from a software background, especially Quality Assurance, this course can be incredibly valuable in helping you elevate your skills and apply them in the AI era.