Weekly two-hour lectures: one live demo, the week's core ideas, and Q&A to close. Slide decks are published here as the semester runs.
Most lectures run to the same shape: one live agentic-coding demo, launched early and left running, two check-ins where you steer it, and Q&A to close. The lecture structure has the detail, including the weeks that break the pattern.
Lectures run Thursdays 11:00–13:00 in the Fulton Muir Building (95), Room 2.02, each teaching week of the semester. They’re captured by the standard ANU AV setup and the recordings land on Canvas — but a good deal of what makes a lecture worth attending doesn’t survive the capture, so see the record before you plan to watch rather than come.
Week 1: Introduction
Welcome to the Agentic Coding Studio
covers: The studio crit model, Agent loops, GitHub Pages deployment, Your participation, Assessment, Communication, Your LLM access, Onboarding, This course is an experiment, The lecture structure, The lecture-demos repo, Crit agents, Software and platforms, Forgotten web, Course overview
Week 2: Specs and self-portrait
The brief poses the problem, the spec is the contract, and your response is a self-portrait of the developer you want to be
covers: Spec-driven development, How browsers work, Claude Code cheatsheet, Task tracking, What's yours in the work, Assessment, The lecture-demos repo, Unsolicited redesign, Assignment 1
Week 3: Backpressure
Harness engineering is the art of giving good negative feedback — the model generates; the harness rejects what's wrong, fast and automatically
covers: Backpressure, Static site generators, Claude Code cheatsheet, Task tracking, GitHub Pages deployment, Assignment 1 retro, Assignment 1
Week 4: Context engineering
The agent's working memory as a designed resource — and, on the web side, the ceiling of static: neal.fun / pudding.cool-style interactive pages
covers: Context engineering, Skills, Claude Code cheatsheet, Task tracking, Your LLM access, Assignment 1 retro, An instrument, Assignment 1
Week 5: Verification
The agent says it's done — but has anyone run it, and do you still understand it? Verification, and the cognitive debt that green checks don't measure
covers: Evaluating agent output, Backpressure, Skills, Assessment, What's yours in the work, A game, Assignment 2
Week 6: Evidence of practice
Read the transcript and the git log closely, reconstruct one consequential decision, and write what you learned into the harness, which the next session loads
covers: Evidence of practice, GitHub Actions basics, Backpressure, The studio crit model, Onboarding, Assignment 2 retro, Assignment 2
Week 7: Full stack
A2 shipped Monday, and the course goes full-stack. Once there’s a server, state has two more places to live; find them, check changes against the running app, and deploy it to Fly
covers: End-to-end tracing, Forms and state, Running in production, Your hosting access, Build the ANU system you wish existed
Week 8: Decision records and persistence
The final project brief is out; then architecture decision records, and how a full-stack app keeps its data across restarts and schema changes
covers: Final Project, Decision records, Persistence, It's alive!
Week 9: Subagents and real-time
What a subagent is and when to use one, then how a change one person makes reaches everyone else's screen
covers: Final Project, All at once
Week 10: Logs, harnesses and multi-user apps
What a deployed app should log, and how two agent harnesses — Claude Code and pi — read the same logs; then what a multi-user app has to handle — telling people apart, what each may do, and two of them acting at once
covers: Final Project, Fly by instruments