Veriwire
← Back to the wire

[One Million AI Engineer Special] How Stanford Teaches AI-Native Engineering — Mihail Eric, Monaco — Latent Space

AchievementResearchNov 8, 2026

In the second run of Stanford's CS146S course, students learn an unfamiliar open-source repo, propose a change, and explain their patch to the maintainer who might merge it. Instructor Mihail Eric, who leads AI at Monaco, discussed the syllabus and his engineering team's practices, including agent workspaces, purchased code review tools, and an on-call agent with access to Datadog and GitHub. The interview marks one million subscribers on the AI Engineer YouTube channel.

Why it matters

Stanford's CS146S course has returned for a second run, redesigned around a question that now faces every software instructor: what homework still teaches anything when a coding agent can complete it in a single prompt. Mihail Eric, who teaches the course and leads AI at Monaco, answered by bringing open-source maintainers directly into the classroom. Students must study an unfamiliar repository, propose a change, and then defend that patch to the maintainer who could actually merge it. The interview, marking one million subscribers on the AI Engineer YouTube channel, also covers how Eric's engineering team operates in practice: Monaco built its own agent workspaces, purchased code review tools rather than building them, and runs an on-call agent with access to Datadog and GitHub. Eric explains his reasoning about which tools a team should own versus buy. The discussion ranges from context engineering in large codebases to background agents like Stripe's Minions and Basis's Authority Map, giving readers a picture of both a changed curriculum and the industry practices informing it.

Receipt № 27971 source · awaiting confirmation ◐

Evidence

1source· awaiting independent confirmation

No score is assigned. Sources and their independence are shown in the citation chain below.

Citation chain · 1 source

01med
latent.spaceBloghttps://latent.space/p/one_million_ai_engineer_special_how_stanford_teaches_ai-native_engineering_mihail_eric_monaco/This source pairs the CS146S syllabus with Eric's own engineering practices at Monaco, including agent workspaces, purchased review tools, and an on-call agent tied to Datadog and GitHub.
PRIMARY
GitHubCompanyStanfordCompanyStripeCompanyDatadogCompanyMonacoCompanyBasisCompanyMihail EricPerson

Related

Nov 8The brain-as-computer metaphor shaped modern AI but may be misleading the consciousness debateNov 8The inventor of the bitter lesson has a bitter lesson for LLM scaling enthusiastsNov 8LLMs are terrible at designing AI systems
Canonical: https://latent.space/p/one_million_ai_engineer_special_how_stanford_teaches_ai-native_engineering_mihail_eric_monaco/