What happened
Mihail Eric, who teaches The Modern Software Developer at Stanford and leads AI at Monaco, brought open-source maintainers into the second run of Stanford's CS146S course. Students must learn an unfamiliar repo, propose a change, and explain the patch to the person who might merge it.
Why it matters
The assignment shifts grading from writing code to navigating a real codebase and defending a change to a maintainer, a skill set that coding agents cannot complete in one prompt, the course appears to assume.
Summaries like this, every morning.
The second run of Stanford's CS146S, taught by Mihail Eric, responds to a concrete problem: a coding agent can finish last year's homework in one prompt. The course now brings open-source maintainers into the loop, asking students to learn an unfamiliar repo, propose a change, and explain the patch to the person who might merge it. That format puts the emphasis on reading and defending code rather than producing it from scratch.
Eric also leads AI at Monaco, where the team built its own agent workspaces, bought its code review tools, and gave an on-call agent access to Datadog and GitHub. His rule for tooling, "You buy to learn, and you build to keep," suggests the course's lessons are meant to survive contact with a real engineering team, not just the classroom.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Kokuyo is consolidating meeting minutes, design documents and tasks into Notion and Notion AI, and linking Cla…
In his Watch Impress column, Nishida Sotokazu says Meta's Muse pushes subscription audits, cheapest electricit…
Testing Rockchip's rknn-llm multimodal demo for Qwen2-VL-2B-Instruct, the author found transformers 5.8.0 thro…
The author built an inventory-demand model with scikit-learn's RandomForestRegressor, saw cross-validation abo…
On the Qualcomm Dragonwing RB3 Gen 2 (QCS6490), YOLOv11-N measured 134.7 ms through default TFLite but 5.03 ms…
Goldman Sachs has barred employees who are not engineers from using advanced AI tools, as reported by Fast Com…
