Skip to main content
A eucalyptus branch entwined with a circular feedback-loop arrow and circuit traces, in a three-ink gold, teal and copper risograph print

Agentic Coding Studio

A course about humans making things for other humans — using AI and other tools.

LLM-based coding agents — tools that run in a loop, editing files, executing code and searching the web to achieve a goal — offer a powerful new workflow for developing software. This is especially true for application domains well-represented in LLM training sets, like the web.

Moving beyond naive “vibecoding”, this course provides a studio-based approach to rapidly prototyping web apps. In the weekly lectures you’ll learn fundamental concepts and practical skills for harnessing an agentic coding workflow. In the weekly studio crit sessions, you’ll demo your work-in-progress prototypes and receive feedback from peers and tutors. By the end of the course you’ll have designed, developed and deployed multiple web app prototypes using a rapid, feedback-driven process.

Learning outcomes

At the completion of this course, you will be able to:

  1. design, build and test full-stack web applications using a rapid-prototyping process
  2. explain how LLM-based coding agents work, including model capabilities, tool use and context management
  3. design and evaluate different LLM agent workflows for software development
  4. apply principles from the scholarly literature to work-in-progress and finalised software prototypes

How the course runs

The course runs in Semester 2, 2026 and is available as COMP4020 (undergraduate) and COMP8020 (postgraduate).

Each week has two components: a lecture covering concepts, tools and techniques, and a studio crit session where you bring your latest prototype and get feedback. Crit groups are kept small (around 20 students each) and led by an experienced tutor, so your work gets real attention every week.

The semester comes in two halves, and this site calls them that throughout. In the static half, up to the week 7 lecture, everything you build is client-side and deploys to GitHub Pages. In the full-stack half after it, your prototypes get a server and a database, and deploy to Fly.io.

The whole course on one timeline: ten weekly crits, three assignments, and the weekly lectures. Hover over (or tab to) any marker for the details, and follow it through to the full brief.

  1. Census date
  2. Last day to drop without failure
  1. Assignment 1
  2. Assignment 2
  3. Final Project (Assignment 3)
  1. C1: Forgotten web (week 2)
  2. C2: Unsolicited redesign (week 3)
  3. C3: Assignment 1 retro (week 4)
  4. C4: An instrument (week 5)
  5. C5: A game (week 6)
  6. C6: Assignment 2 retro (week 7)
  7. C7: Build the ANU system you wish existed (week 8)
  8. C8: It's alive! (week 9)
  9. C9: Crit the critics (week 10)
  10. C10: Fly by instruments (week 11)
  1. Setup drop-in
  2. Redeemable crit
  1. Week 1: Introduction
  2. Week 2: Specs and self-portrait
  3. Week 3: Backpressure
  4. Week 4: Context engineering
  5. Week 5: Verification
  6. Week 6: Evidence of practice

assignments weekly crits one-off sessions lectures enrolment dates today

Prerequisites and enrolment

Because it’s a new course, all enrolments go through ANU’s permission code application — even if you satisfy the prerequisites.

If you’ve applied and haven’t heard back after a week, email student services to follow up.

Course materials

Lectures

Weekly lecture slide decks.

Crits

Weekly studio crit sessions.

Assessments

Assignments and project information.

Topics

Browsable content chunks covering individual concepts, tools, and activities.

Admin

Course policies and administrative information.