<div dir="ltr"><p data-sourcepos="1:1-1:1091">Azure AI Connect stands as a vibrant series, meticulously crafted for Microsoft Professionals, esteemed MVPs, and passionate Technology Enthusiasts, all converging to explore the expansive Azure AI Ecosystem. We find ourselves at the dawn of the AI Application era, a period of unprecedented growth where collective learning and evolution are paramount. The sheer magnitude of AI Technology advancements is staggering, and Azure emerges as a premier ecosystem, empowering a diverse spectrum of Applications, groundbreaking Innovations, and seamless Integrations. With its rich array of tools and capabilities, Azure stands ready to fuel the next wave of technological progress. This series is your gateway to unraveling the intricacies of Azure AI, offering invaluable insights and practical knowledge. By participating in these sessions, you'll embark on a journey of continuous learning, evolving alongside seasoned Professionals and recognized Experts, ensuring you remain at the forefront of this transformative technological revolution. Join us, and be part of shaping the future of AI.</p></div>
<p></p>
<p></p>

Azure AI Connect - Global AI Bootcamp

Room 1

Get started using Microsoft's Semantic Kernel for .NET to connect to your OpenAI account hosted on Azure. Explore the Semantic Kernel API and learn how to build .NET apps which utilize OpenAI.

Using Semantic Kernel in .NET with Azure AI

As vector search and RAG become mainstream for GenAI use cases, we’re looking ahead to what’s next. Generative Feedback Loops (GFL) represent the next evolution in AI where the AI outputs are improved over time through a cycle of feedback and learning based on the production data. In this talk, we’ll break down how this works, and you will learn how to build a sample AI app with feedback loops in .NET using Azure Functions, Neon serverless Postgres, and Azure OpenAI. 

Generative Feedback Loops (GFLs) in action

What if I told you that you could use GPT-powered natural language to investigate and respond to security incidents, threats, and vulnerabilities facing your organization right now? This presentation on Microsoft Copilot for Security will showcase its advanced integration with Microsoft 365 Security technologies. This innovative tool leverages AI to enhance threat detection, automate responses, and provide comprehensive insights into security incidents. By seamlessly integrating with Microsoft Defender, Azure AD, and other 365 security services, Copilot for Security elevates organizational protection and streamlines security operations, ensuring robust defense against evolving cyber threats.

Microsoft Copilot for Security: Use AI to defend!

Did you know that you can combine various AI services such as Open AI, Azure Open AI and Hugging Face in your own custom application?

In this session you will learn how you can build your own CoPilot experience using the Semantic Kernel which is released as an open source project. 

I will demonstrate how you can use skills, memories, connectors and plugins in order to enhance the experience for your users.

This session is a developer focused session meaning that you do not need to have any AI knowledge in order to benefit from the examples demonstrated in the session.

How to enhance your own AI services using Semantic Kernel

Discover Firebase Genkit in this introductory workshop. We'll explore how this versatile framework empowers developers to build AI-driven applications with ease. Learn to deploy Genkit's open-source applications for Node.js and Go, integrate with generative AI models, and run independently of Google services. Perfect for those looking to streamline their AI development process. We will talk about Github, Azure and Google Models too.

Introduction to Firebase Genkit

This presentation explores the transformative capabilities of Retrieval-Augmented Generation (RAG), a cutting-edge framework that integrates retrieval systems with generative language models. Tailored for researchers and practitioners, the session provides a detailed examination of RAG's architecture, demonstrating its ability to overcome challenges such as static knowledge and limited contextual understanding in traditional models.

The talk includes an in-depth walkthrough of RAG's implementation pipeline, covering techniques for efficient information retrieval, integration with generative models, and optimising responses for diverse applications such as question answering, summarisation, and dialogue systems. Attendees will gain a robust understanding of how to build and deploy RAG systems effectively, bridging the gap between retrieval and generation for state-of-the-art performance in real-world scenarios.

Demystifying Retrieval-Augmented Generation (RAGs)

Discover how Microsoft Azure AI Agent capabilities and Azure AI Foundry are revolutionizing AI-driven automation. This session will introduce the fundamentals of AI agents, explore Azure AI Foundry, and demonstrate how multi-agent automation enhances productivity across various use cases. We'll dive into practical implementations using Autogen, showcasing how multiple specialized agents can collaborate seamlessly. Join us for an insightful session with a live demo, where you'll learn how to build intelligent, automated workflows with Azure AI.

Unlocking AI powered automation with Microsoft Azure AI Agent

Artificial Intelligence (AI) has rapidly evolved, promising transformative advancements. However, as AI systems become increasingly sophisticated and integrated into our daily lives, so do the security risks associated with them. This talk delves into the realities of AI security, beyond the hype and promises. We will explore the vulnerabilities inherent in AI systems, including data privacy breaches, adversarial attacks, and model poisoning. The presentation will also discuss the ethical implications of AI, such as bias and discrimination, and how they intersect with security concerns. By the end of this talk, attendees will gain a comprehensive understanding of the challenges and best practices for securing AI systems, enabling them to build and deploy AI solutions responsibly.

Between the main points of the talk we can highlight:
-The evolving threat landscape(Adversarial attacks,data privacy and security)
-Ethical implications of AI security
-Securing AI systems: Best practices and emerging technologies
-The future of AI security

Beyond the hype: The reality of AI security

Hands On: Copilot for Security

Discover the core principles and best practices of deploying AI solutions in the cloud. This session will dive into reference architectures, covering key components like data pipelines, model training, deployment, and scalability. Learn how to leverage cloud services for efficient, scalable, and robust AI implementations. Ideal for architects and developers.

Reference Architecture of AI in the Cloud

25 years after Agent Smith coined “Never send a human to do a machine’s job”, this futuristic idea seems closer than ever...

Not so long ago, AI was known for its ability to excel at specific tasks (“cat vs dog”), earning it the reputation of a specialist. However, with the advent of Large Language Models like OpenAI’s GPT, AI has transformed into a generalist, capable of writing naturally, answering questions, and even generating code. Despite facing limitations such as hallucinations, knowledge cutoffs, and struggles with basic math, these AI systems have opened up a world of possibilities.

Today, the next frontier in AI development are AI agents. By combining the power of LLMs with advanced capabilities such as planning, reflection, and tool use, AI agents are becoming the “jack-of-all-trades” in the tech world. Drawing inspiration from human cognition, AI Agents can accomplish complex goals, thinking and acting much like us humans. Join us as we explore agentic workflows and discuss the role of developers in guiding AI agents and shaping the future of AI. We’ll get familiar with the world of AI agents, and understand their potential to revolutionize the way we work and interact with technology, ultimately allowing humans (and developers 😉) to focus on the bigger picture.

The Rise of AI Agents

In this session i will be showcasing you how to create your first AI based application using RAG model

Build Your First AI Application with RAG

The most popular AI demo is the AI chatbot that responds to questions asked by a human. This is a valid and instructive scenario but we aspire to go beyond the basics in this talk.

Imagine an AI that not only answers your questions but also actively works towards goals you've assigned to it. In this session, we'll explore the building of an AI Agent – an autonomous actor that can execute tasks and achieve objectives on your behalf.

We'll start with a basic C# chatbot built using Microsoft's Semantic Kernel framework. Then we'll transform it into a simple yet powerful AI Agent by giving it access to some tools and assigning it a goal to accomplish using those tools.

The end result is the core of an AI Agent - a tool that can do work on your behalf without you being there. We will also compare this to a Copilot since there's substantial overlap.

By the end of this talk, you'll understand how to approach building goal-driven AI solutions using Semantic Kernel and (hopefully!) be inspired to apply these techniques to your own projects.

This talk assumes user-level familiarity with LLMs like ChatGPT or Microsoft Copilot and basic prompt engineering concepts. Anything else will be explained. 

From Chatbot to Goal-Oriented AI Agent Using Semantic Kernel

Do you need to search your data based on meaning instead of matching by keywords or phrases? Do you need to match the data in a language different from the query term?

Semantic search or search based on the meaning analyzes the context and intent behind the query term to provide relevant results. Using Vector embeddings, the data structure behind semantic search, you can supercharge your search to include text, images, and other types of data. With vector databases you can store and index vector embeddings and provide similarity search over these embeddings.

In this session, we will explore the building blocks of semantic search and explore vector embeddings and similarity metrics. You will learn how to generate embeddings with large language models using OpenAI. I will also explain how to store, index, and query vector embeddings using the vector data type in Azure SQL Database or with the upcoming Microsoft SQL Server 2025. Finally, you will see how optimize the queries with the DiskANN vector index.

Join me for a demo-rich session and learn how to implement semantic search in with Azure SQL Database.

Supercharged Search with Semantic Search and Vector Embeddings

Data-Centric AI: Accelerating Success Through Modern Data Products

Target Audience
- Data Leaders & Architects
- Data Engineers & Platform Engineers
- ML/AI Engineers & Data Scientists
- Analytics Engineers & Practitioners

Abstract
While organizations rush to advance their AI models, research shows that "reducing the technological gap alone is not enough to ensure success in AI projects" [The Data Death Cycle, 2024]. The key to accelerating AI success lies not in model optimization, but in data excellence. This session reveals how a data-centric approach, powered by Apache Iceberg and modern data architectures, dramatically improves AI systems by ensuring complete, consistent, and curated datasets from the start.

Overview
AI and analytics initiatives demand high-quality data, yet traditional model-centric approaches often overlook this fundamental requirement. We'll explore how data products, implemented through a combination of data mesh and data fabric patterns, provide the systematic data excellence that AI requires. Learn how Apache Iceberg's lakehouse architecture eliminates costly ETL while enabling "git-like" version control through metadata catalogs like Polaris and Nessie, providing comprehensive write-audit-publish capabilities for data changes.

Through real-world examples and architectural patterns, we'll demonstrate how organizations can:
- Accelerate AI success through systematic data excellence rather than just model optimization
- Create trusted data products that ensure quality, completeness, and consistency
- Implement efficient data integration without expensive ETL processes
- Enable version control and auditability for data changes
- Balance centralized governance with domain agility

Key Takeaways

1. Data-Centric Advantage
- Why focusing on data quality accelerates AI success more effectively than model optimization
- How systematic data excellence reduces the 80% of time data scientists spend on data preparation
- The critical role of complete, consistent, and curated datasets in AI/ML success

2. Modern Data Products
- How data products enable systematic management of data quality
- Why combining data mesh and data fabric creates the ideal architecture for AI-ready data
- Patterns for implementing data products that ensure quality, governance, and reliability

3. Technical Foundation
- How Apache Iceberg enables efficient data integration without costly ETL
- The role of metadata catalogs in providing git-like version control for data
- Practical patterns for implementing write-audit-publish workflows for data changes

4. Implementation Path
- Steps for transitioning to a data-centric approach
- How to begin implementing data products in your organization
- Methods for measuring and demonstrating success

Whether you're struggling with AI initiatives or looking to accelerate existing programs, this session provides practical insights into building the data foundation that modern AI demands. You'll learn why data-centric approaches succeed where model-centric efforts fail, and how to implement the technical architecture that makes it possible.

Join us to discover how combining data-centric thinking with modern technologies like Apache Iceberg can transform your organization's ability to deliver successful AI initiatives. Leave with concrete steps for implementing data products that provide the complete, consistent, and curated data that AI systems require.

Data-Centric AI: Accelerate Success with Apache Iceberg Data Products

Discover how to use Azure OpenAI real-time APIs with your applications to build low-latency, AI-powered voice agents. This session will demonstrate how to create interactive experiences that process and respond to audio input instantly over VOIP and telephony. The talk will include a case study and a live coding demo to illustrate the concepts.

Build low-latency AI voice agents using real-time GPT models

In this session, we’ll explore how Azure OpenAI and Azure Search can improve your document processing. Learn how Azure OpenAI can help with tasks like extracting and summarizing content from documents, while Azure Search allows you to quickly find and analyze information. We’ll look at real-world examples showing how these tools work together to automate document handling, save time, and make it easier to find important insights. By the end of the session, you'll have a clear understanding of how to use these tools to streamline your document workflows.