The Foundations Series: Building Reusable Skills
Session 2 — Building Reusable Skills
The Public Sector has deep analytical expertise. How could AI help us extend what that expertise makes possible—investigate more questions, respond sooner to emerging issues and provide stronger evidence for policy and services?
In this hands-on Foundations Series workshop, we’ll explore how reusable AI Skills can support analytical work and make established approaches easier to share.
Working alongside colleagues and coaches from government, academia and industry, you’ll practise turning your knowledge into instructions and checks that AI can follow, while keeping your judgement at the helm.
Together, we’ll:
Work through a shared data-preparation example using synthetic data, identifying where AI can reduce repeated effort.
Capture analytical know-how in reusable Skills, drawing on existing data dictionaries, code and reference materials.
Set clear points for AI to ask questions, confirm assumptions and return work for human review.
You’ll leave better equipped to recognise where AI can help, test and improve its contribution, and develop Skills you can adapt to your own work.
The opportunity is for one team’s progress to become another team’s starting point. Reducing repeated effort can give analysts more time to examine the evidence, test assumptions and explore questions that might otherwise go unanswered.
This workshop supports reusable AI assets and capability that can be shared across agencies, and eventually with the not-for-profit sector.
Please bring your own personal laptop (as we will be working off departmental systems) and aim to arrive at 9am to check tech is working before we begin at 9:30am.
Building shared foundations for public impact
The Foundations Series brings government, academia, industry and the not-for-profit sector together to learn by doing — small, practical projects that build and share reusable AI skills, code and methods on public GitHub, so small teams and public impact entrepreneurs can do more with open data and expertise for public benefit. If you’d like to suggest a future session topic or join as a participant, coach or partner, contact the Series Convenor, Simon Nash.
About the workshop team
Gayatri Kotnala - Foundation Series Coach
Gayatri is the founder of Pypline, a Canberra-based firm that designs AI systems and ways of working around human judgement, working with government and higher education clients. She has experience across operational leadership and strategy consulting. Her work focuses on helping teams get reliable results from AI, including building reusable skills and custom agents in Microsoft Copilot Studio, Claude and ChatGPT.
Simon Nash — Foundation Series Convenor
Previously Director of Data Strategy and Governance and Business Development Manager in Treasury's Macroeconomic Analysis and Policy Division, and Co-Director of Finance's Acceleration team in the AI Delivery and Enablement Division. He convenes the Foundations Series, focusing on cross-agency and cross-sector collaboration for public impact.
AI CoLab events are intentionally open and collaborative. We capture video images, 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.