AI for the Systems You Inherited (AI CoLab)

​AI for the Systems You Inherited

​Every APS agency runs on systems somebody else built. Some were built decades ago. The people who wrote them have moved on, the documentation is thin or wrong, and there is constant pressure to modernise. For many teams, the hardest part is not replacing these systems. It is understanding them well enough to know where to begin. AI is starting to shift that.

​In this hands-on workshop, we will explore what is actually possible with AI in legacy modernisation, and what still needs human judgement. The public conversation is often split between hype — “AI will translate your COBOL by Friday” — and dismissal — “AI engineering is slop”. Neither is useful on its own. We will look at what practitioners and researchers are learning, and test where AI can genuinely help people reason about old systems, unclear documentation and messy business rules.

​The session is built around three disciplines: Design — understanding the problem and the users; Architecture — mapping the system and choosing what to modernise first; and Development — translating code and proving it still works. For each, we will ask the same practical question: where can AI help, and where do good designers, architects and engineers remain essential?

​The first hour is discussion and shared sense-making. The second hour is hands-on, working in small groups with a realistic legacy codebase connected to a business problem. No matter your discipline, there is a way in.

​What to bring: To contribute to the workshop portion of this session, you will need to bring a laptop with access to an AI chat (MS copilot, ChatGPT, Claude etc), and be able to download files onto your laptop

​Facilitators

​Dr Rishni Ratnam (CEO, MXA Consulting), Dimi Dimitrovski (CIO, Fair Work Ombudsman) and Justin Butcher (Chief Architect, Department of Finance).

​All three have spent years working with exactly these kinds of systems, inside and alongside the APS.

​​​AI CoLab events are intentionally open and collaborative. We capture photos, audio, and AI-generated transcripts so we can remix key insights for the AI CoLab Alliance community. This is always done in line with the Charter’s values; transparent, ethical innovation and knowledge sharing to accelerate collective learning (see join.aicolab.org). By participating in an AI CoLab event, you agree to your contributions being captured and shared in accordance with this policy.