Post Tech Summit Workshop 2026: Build an AI agent you can actually trust with Stefania Druga
Description
Build an AI agent you can actually trust
This is the hands-on practical workshop.
In small teams, you'll work with a real dataset and build your own version of a trustworthy AI agent: getting it to form a testable idea, checking that idea against evidence it hasn't seen, and recording why it made the call it did.
Partway through, new evidence will overturn an early decision. You'll see the difference between an AI that just remembers a summary and one that keeps an auditable record of what changed and why, and you'll practise making that call yourself.
You'll leave with a repeatable workflow for building AI agents you can trust, plus a simple one-page template for documenting how your own agent makes decisions.
Who it's for:
Developers, product and digital teams, and founders who want hands-on practice, not just the concept.
Best suited to people who are comfortable with the idea of working with an AI agent directly or who plan on seeing Stefania's keynote at Tech Summit the day prior (www.techsummit.nz).
Things to note:
Please ensure you purchase the correct ticket. You must either have a Tech Summit 2026 ticket, be a current Canterbury Tech member, and if not, please purchase a Non-Member ticket.
Members must have a valid membership at the time of booking and on the day of the event. This will be validated at each point and we reserve the right to invoice the event at the full, non-member rate in the event that membership has lapsed.
Please note that registering for this event will be taken as acceptance that your image, name, or voice may be included in a recording or photos of proceedings that will be published after the event.
By registering for a Canterbury Tech event, you are agreeing to abide by our Code of Conduct, Terms and Privacy Policy.
Lineup
Stefania Druga
Stefania is a Research Scientist at Sakana AI in Tokyo, where she works on culturally-aligned AI systems and AI-assisted scientific workflows. She was previously at Google DeepMind, working on novel multimodal AI applications. She holds a PhD from the University of Washington and a master’s from MIT, where she built Cognimates, the first open-source platform for K-12 AI education. Stefania has been researching AI literacy and human-AI interaction since 2015, with a focus on how diverse communities engage with and shape AI systems. She co-founded HackIdemia, a global STEM education NGO with an active Singapore chapter, and leads the AI Tinkers community in Tokyo.
Tickets for good, not greed Humanitix dedicates 100% of profits from booking fees to charity
Register
Register
More events from this host
