How Stanford Teaches AI-Native Engineering — Mihail Eric, Monaco
· Source: Latent Space
Mihail Eric, a professor of Stanford’s “The Modern Software Developer” course and AI lead at Monaco, explained how the CS146S program is structured. The ten‑week course blends theory with hands‑on coding agents. It includes nine assignments and a final project, and has evolved so that students work directly on open‑source projects, submitting patches that are reviewed by the actual maintainers. This dynamic forces students to use agents that generate features quickly while also understanding the project’s context, validating the need for change, and anticipating potential breakages.
At Monaco, the team evaluated various review tools and decided to build their own “agent workspaces”—environments where agents can reproduce bugs and test fixes before opening a pull request. The group’s rule is to buy solutions to learn from and then develop internally what is needed to maintain control.
This initiative illustrates how software engineering education is adopting generative AI as a core curriculum component, preparing future developers to work with tools that automate much of the development cycle. Integrating agents into real workflows also raises questions about change management and accountability in open‑source projects—critical issues for software quality and sustainability.
Read the original article on Latent Space
This summary is an informational synthesis produced by dataqbs.com. All rights to the original content belong to its author and the cited media outlet. We act solely as curators of technology news and claim no authorship.