Microsoft Reactor San Francisco

AI agents are transforming applications with autonomy and intelligence, but how do you keep them secure? This session is a practical deep dive into securing agentic AI on Azure, covering real-world threats, security best practices, and hands-on demos.

Follow along with the content here:

This module provides engineers with the skills to begin building agents with Azure AI Agent Service.

Learning objectives
By the end of this module, you'll be able to:

• Describe the purpose of AI agents
• Explain the key features of Azure AI Agent Service
• Build an agent using the Azure AI Agent Service
• Integrate an agent in the Azure AI Agent Service into your own application

-----------------------------------------------
Follow along with the content

Microsoft Purview Data Security Posture Management (DSPM) for AI helps organizations monitor AI activity, enforce security policies, and prevent unauthorized data exposure. Learn how to configure DSPM for AI, track AI interactions, run data assessments, and apply security controls to reduce risks associated with AI usage.

***

Follow along with the content:

In this new era of AI agents and nascent robotics interactions, creating truly engaging and intelligent experiences requires real-time, low-latency communication, adaptive behavior, and seamless multimodal integration across text, speech, and vision. The GPT-4o real-time API in Azure OpenAI Service unlocks new possibilities for robotics developers, enabling robots to process natural language and speech, interpret images and context, and generate dynamic responses with minimal latency.

This session dives into crafting AI experiences that feel truly alive across both physical and virtual environments. Attendees will explore how to infuse AI with personality, speech, animations, and facial expressions—transforming it into an interactive agent capable of engaging its surroundings with dynamic, lifelike qualities. Additionally, we'll discuss using virtual 3D bodies as a test bed for robotics applications.

In this talk, we’ll explore how real-time capabilities can power interactive backends for robotic and immersive applications, highlighting both their strengths and limitations. We’ll discuss the challenges encountered along the way, the feasibility of using this technology today, and whether it meets the demands of real-world applications or still needs additional capabilities.

Through live demos and practical examples, you’ll see how AI combined with 3D embodiments can revolutionize entertainment, education, customer service, and beyond. Whether you’re an AI engineer, innovator looking to push the boundaries of human-AI interaction, or a roboticist interested in bringing robots to life with LLMs, this session will provide actionable insights to take your projects to the next level.

We'll cover two technology stacks:

Three.js, React, Ready Player Me, Blender, Azure GPT-4o Realtime API
Unity, SALSA LipSync Suite, Eleven Labs Text-to-Speech, Ready Player Me

Join us to discover how Azure OpenAI is shaping the future of real-time, interactive AI experiences!

Follow along: