Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the emerging field of mechanistic interpretability and its practical applications for AI engineers in this 21-minute conference talk. Learn how reverse engineering neural networks provides direct, programmable access to internal model neurons, enabling more precise AI steering, enhanced guardrails, and innovative user interfaces. Discover why interpretability research is transitioning from academic curiosity to real-world implementation, making it an essential skill for modern AI development. Gain insights from an applied researcher's perspective on translating frontier research into practical solutions, and understand how recent AI developments are reshaping professional roles to blur traditional boundaries between engineering, research, invention, design, and entrepreneurship.
Syllabus
Why you should care about AI interpretability - Mark Bissell, Goodfire AI
Taught by
AI Engineer