[One Million AI Engineer Special] How Stanford Teaches AI-Native Engineering — Mihail Eric, Monaco
TL;DR
What do you assign when a coding agent can finish last year’s homework in one prompt? In the second run of Stanford’s CS146S, Mihail Eric brought open-source maintainers into the course. Students have to learn an unfamiliar repo, propose a change, and ... What do you assign when a coding agent can finish last year’s homework in one prompt?
Nauti's Take
For a small team, the useful experiment is straightforward: give an agent an unfamiliar repository and score its architecture explanation, proposed change, and tests before looking at the final patch. That reveals whether the workflow produces real understanding or merely plausible code.
Stanford’s course is a promising model to watch, but the available report does not yet show enough detail about grading or student outcomes to copy it wholesale.
Video
Sources
