In the rapidly evolving landscape of Artificial Intelligence, the ability to execute code securely in a performant way is paramount. This presentation introduces Azure Container Apps dynamic sessions as a robust solution for running code within an AI system. By leveraging isolated, sandboxed environments, the system can confidently execute potentially untrusted code, including user-provided scripts or code generated by Large Language Models (LLMs) while mitigating risks and ensuring the integrity of the system. This session will equip you with the knowledge to confidently integrate secure code execution into your AI solutions, minimizing vulnerabilities and maximizing trust.