Building secure AI applications in practice

Building AI features today is easier than ever. With LLM APIs and AI coding tools, teams can quickly prototype and ship AI-powered apps. But when these systems move closer to production, many hidden challenges start to appear.

• How do you handle untrusted user input?
• How do you prevent prompt injection or poisoned data?
• What happens when an AI model has access to internal systems or sensitive data?

In this session, we’ll walk through the design of a real AI application and highlight the common pains teams face when building AI systems in practice. We’ll discuss practical ways to design safer architectures, add guardrails, and reduce risks while still moving fast with AI.

Agenda:
6.15 - 6.30pm: Registration and welcome
6.30 - 7.10pm: Building secure AI applications in practice
7.10 - 7.15pm: Q&A and networking

Meet the speaker:
Duc Nguyen (Mata) is a Senior Consultant with experience in software engineering, DevOps, and AI. He has worked as an AI engineer across domains including banking and hiring platforms, building scalable systems and developer tools.

Venue:
15 Floor, Pearl Building, 5 Le Quy Don, Xuan Hoa Ward, HCMC, Vietnam
(Upon arrival, find the visitor parking at level B1 and take the emergency exit to go to level 1, register your name with the security at the lobby, then take lift to the 15th floor)

To register for this in-person talk, simply click "Attend" here on Meetup and you will receive a confirmation email.

This is a free event. Food and drinks will be served.

*Disclaimer: Photographs, videos or recordings may be taken and used by Thoughtworks at this tech talk. Thoughtworks may use these photos, videos or recordings in public transmissions such as in media accounts, digital platforms, brochures, websites and other publication materials. If you have any concerns with being photographed, filmed or recorded, please let the onsite photographer, videographer or event organizer know. For further information, please refer to Thoughtworks' Privacy Policy listed in our website: www.thoughtworks.com