TL/IRL: AI for Security & Security for AI [Georgian x AISecEng]

AI is a powerful security tool, but it's also a new attack surface! 🔐 Open models can help hunt for vulnerabilities in your code and dependencies, but they also bring risks of their own, from missing guardrails to a compromised supply chain. The AI products we build on top of them need permissions and boundaries that hold up in production, especially as agents adopt skills that can introduce malicious instructions or compromised dependencies. So what does it look like to use AI to find threats? And what does it take to keep the AI you ship from becoming one? Join us for a night of talks from practitioners tackling both sides of the problem, hosted by Georgian's Transferred Learnings and AISecEng. What we'll cover Open models as threat hunters: how well they work, what they cost and where they fall short Securing AI products: permissions, guardrails and boundaries that hold up in production AI for security work: vulnerability discovery and when to buy vs. build Agent skills security: malicious instructions, compromised dependencies and practical defences Agenda 5:30 – 6:00 PM: Welcome & food 🍕 6:00 – 7:15 PM: Speaker talks 🎬 7:15 – 8:30 PM: Networking 💬 ​Speakers Ben Wilde, Head of Thesis and Innovation @ Georgian Ganni Galea Curmi, AI Engineer @ Georgian Miguel Young, Staff Enterprise Security Engineer @ ClickUp Nipun Gupta, Founder & CEO @ Optimus Labs About the Hosts Georgian Transferred Learnings TL/IRL is an in-person AI talk series by Transferred Learnings — Georgian's technical community for AI builders — designed for practitioners (researchers, engineers, founders) who are actively shipping AI solutions and want to stay ahead of the field. Built around the spirit of Applied AI, each session features the latest trends and practical insights from builders actually moving the needle. Conversations assume a foundational ML background and hands-on experience, so if you're already building, you'll feel right at home. Interested in joining? Submit your application here! AISecEng Toronto The AI Security Engineers Community (AISecEng) Toronto is a local, practitioner-led chapter dedicated to the emerging field of AI and machine learning security. It brings together security engineers, researchers, and developers to share practical strategies for securing AI technologies.