In the second run of Stanford’s CS146S, Mihail Eric — who teaches The Modern Software Developer at Stanford and leads AI at Monaco — redesigned assignments to reflect the presence of AI coding tools. Students are required to learn an unfamiliar open‑source repository, propose a change, and explain the patch to the person who might merge it (the maintainer).
Why this matters
The central question the course addresses is: what tasks do we assign when a coding agent can reproduce last year’s homework from a single prompt? The syllabus emphasizes skills beyond raw code generation: understanding context inside a foreign codebase, identifying design decisions, and communicating changes effectively to reviewers.
The interview and demonstrations
To mark one million subscribers on the AI Engineer YouTube channel, Mihail Eric walks through the course syllabus, the ten‑week assignments, and which lessons held up after being stress‑tested with his engineering team at Monaco. The full interview and accompanying screenshares are available on YouTube.
Monaco’s engineering choices for agents
During the interview, Eric explains how Monaco built agent workspaces, which code‑review tools the team purchased, and how they provisioned an on‑call agent with access to operational systems such as Datadog and GitHub. He also discusses how his team decides which tools to buy and which to build in‑house.
Show notes and referenced materials
Items mentioned in the interview include:
- Stanford CS146S — the ten‑week syllabus and assignments
- The Emperor Has No Clothes — Mihail builds a coding agent from scratch
- Demystifying Claude Code — tracing a real agent harness
- Monaco’s agent workspaces — how the team sets up a workspace for each agent
- How to Kill the Code Review — an essay on what happens when code generation outpaces review
- Basis’s Authority Map — how agents distinguish current behavior from earlier design decisions
- Stripe’s Minions — a look inside Stripe’s background coding agents
- AI Engineer CODE Summit — the conference mentioned at the start of the interview
Interview structure — timestamps
Key segments in the YouTube interview (timestamps):
- 0:00 — One million AI Engineer subscribers and the CODE Summit
- 0:51 — Mihail Eric and Stanford CS146S
- 5:26 — What junior engineers need to learn now
- 10:02 — Open‑source maintainers review the students’ code
- 15:04 — A tour of the ten‑week syllabus
- 24:04 — The loop, tools, and harness inside a coding agent
- 32:18 — Context engineering for a large codebase
- 38:15 — Skills that make you understand the diff
- 46:42 — The Monaco.com dispute and the airplane banner
- 50:23 — What Monaco does for sales teams
- 59:21 — Inside Monaco’s engineering team
- 1:01:40 — “You buy to learn, and you build to keep”
- 1:05:24 — Making repos agent‑ready and closing the on‑call loop
- 1:12:37 — When code review becomes the bottleneck
- 1:20:37 — Security, dependencies and permissions
- 1:30:13 — Background agents and Stripe’s Minions
- 1:34:16 — MCP portals and why infrastructure converges
- 1:41:22 — Building the software factory
- 1:45:24 — What work is left for developers?
- 1:52:16 — From personal adoption to team autonomy
Conclusions
Stanford’s CS146S and the assignment design Mihail Eric describes push students to acquire collaborative, engineering‑level skills that remain relevant as generative tools proliferate: rapidly understanding unfamiliar repos, producing diffs that can be reasoned about, and explaining changes to maintainers. Monaco’s experience shows practical steps for integrating agents into operational workflows, combining per‑agent workspaces, monitoring access, and permissioning to close the on‑call loop.
For a deeper technical walkthrough and the complete syllabus, watch the full interview with screenshares on the AI Engineer YouTube channel.
