<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 2

We hear about the dangers of Generative AI. Generative AI can exclude people. The system can generate bias and it's possible to strengthen the current stigma people from diverse backgrounds may have. But for this talk, let's reverse the question: Can Generative AI make the world more inclusive, in specific for people with disabilities?

As a tech-optimist I would say yes. But you don't need to believe me. Instead of a blind believe in the yes, I give you some interesting cases where Generative AI limits bias and makes the world more inclusive.

- For people without formal education, Gen AI can explain theories;
- For people with dyslexia, Gen AI can write stuff in better wordings and grammar;
- For people with autism, Gen AI can help them to understand social situations better;
- For people with a visual impairment, Gen AI can help with reading text;

In this talk we are going to explore some use-cases how Gen AI can make the world more a more inclusive and better place for everyone, in specific for people with disabilities.

Can Gen AI make the world Disability inclusive?

AI in Education: Nonlinear and Exponential

1) Personalized Learning (Content for One): Tailoring educational content to align with the unique preferences and learning styles of each individual student.
2) Real-Time Feedback: Empowering students with instant insights through AI, ensuring they grasp concepts accurately during the learning process.
3) Dynamic Pathways: AI creating adaptive learning pathways or course plans that evolve in real-time, accommodating changes in goals or external circumstances.

AI in Education

In the rapidly evolving landscape of artificial intelligence, Large Language Models (LLMs) are proving to be both revolutionary and dangerously vulnerable. While their transformative potential is undeniable, their deployment creates new risks and amplifies existing threats. Alarmingly, current enterprise security tools are unequipped to address these vulnerabilities, leaving organizations exposed. As a result AI systems are either going live without proper security assurances or being blocked altogether, stifling innovation and adoption.

In this talk, Dr. Peter Garraghan takes the audience on a deep dive into the underbelly of AI vulnerabilities. Drawing on over a decade of research, Dr. Garraghan will demystify how attackers exploit weaknesses in AI to disrupt services, leak sensitive data, and reverse-engineer proprietary capabilities. He will also highlight how the unique, dynamic behavior of AI amplifies these risks, making them particularly vulnerable to sophisticated attacks that can be triggered through seemingly benign user interactions, turning ordinary usage into a potential threat vector.

Through a live demonstration, attendees will witness how adversarial attacks can be launched quickly and cheaply, highlighting the need for AI safeguards. Dr. Garraghan will introduce practical guardrails, emerging AI security tooling, and best practices to mitigate AI risk, empowering organizations to adopt AI safely and securely.

Can My AI Be Hacked?

We are moving on from building just intelligent systems to creating responsible, ethical, and impactful solutions. Join us for a hands-on session on building ethical and safe AI solutions using Azure AI Foundry. Let's cut through the jargon and discuss practical ways to make your AI projects more responsible. In this session, you'll learn:
	• The basics of Responsible AI and why it matters
	• How to use Azure AI Foundry's tools for safer AI development
	• Practical tips for content moderation in AI applications
	• Real-world techniques for ethical AI governance

The session is intended for developers, data scientists, and tech leaders who want to create AI that's not just smart, but also trustworthy. Let's see how it's done. 
Key takeaways:
	• Understand what Responsible AI really means in practice
	• Learn to use Azure AI Foundry's safety features effectively
	• Master practical content moderation techniques
Implement ethical AI governance in your projects

Building Trustworthy AI: Practical Approaches with Azure AI Foundry

Learn how Azure OpenAI empowers you to scale your LLM deployments, ensuring they are cost-effective, high-performance, and ready for the real world. Discover how to navigate critical metrics like TPMs, RPMs, and PTUs, tackle latency challenges, and handle bursty traffic with Azure API Management (APIM). We’ll share practical strategies for controlling costs, enabling multi-region scalability, and using Azure’s capacity planning tools to meet your growing needs. Whether you're scaling a chatbot, automating document workflows, or building enterprise AI applications, this session provides the roadmap to turn your LLM aspirations into operational success.

Discover powerful frameworks that take enterprises from concept to implementation—whether on the cloud, on-prem, or in a hybrid environment.
Ready to take your AI initiatives to the next level? Join this session to unlock the secrets to building enterprise-grade LLM deployments that are truly game-changing.

Accelerating AI Success: Leveraging Scalable and Cost-Efficient LLMs with Azure

In the rapidly evolving world of AI, choosing the right tools can make or break your project. This talk dives into the pros, cons, and performance of Azure AI Language versus popular open-source NLP solutions (Hugging face, Haystack, SpaCy, etc). From scalability and integration to cost and customization, discover which approach aligns best with your goals. Whether you're a cloud enthusiast or an open-source advocate, this session will provide valuable insights to empower your AI strategy.

Battle of the titans: Azure AI Language vs. OpenSource solutions

AI has transformed various industries, including DevOps, by preventing service disruptions, improving security, and enabling anomaly detection. In this talk, you'll learns some of the best ways to apply AI to manage and secure Azure infrastructure, from code generation to policy enforcement to remediation. 

We'll walk through real code examples and applications, that will enable you to supercharge your cloud ops and security through AI.  

Don't miss out on this opportunity to stay ahead in the AI revolution.



Elevating Cloud Security with AI: A DevSecOps Revolution

In this session we will learn how to boost your copilot studio generative AI responses by connecting it to Azure Search through Azure Open AI. In this way, we can create our own index configuration in Azure AI Search, using semantic ranking to prioritize the content that is more relevant to our users. 

Boost Copilot Studio using Azure AI Search

This session will guide participants in enhancing and integrating an application with Azure services. Key tasks include configuring and integrating the app with Azure Kubernetes Service (AKS). Additionally, participants will expose OpenAI via API Management, update Docker images, and build a Copilot App using Semantic Kernel Tools.

Build Intelligent Apps with Microsoft's Copilot stack & Azure OpenAI

AI technologies are increasingly integrated into many aspects of daily life, especially for developers. However, this integration has not yet consistently resulted in inclusive and accessible products and services. In this presentation, I will explore how Azure AI is leading the way in achieving accessibility at scale. Through two compelling use cases, I will demonstrate how these technologies can be seamlessly incorporated into design processes to enhance services. First, discover a news site enhanced with Azure’s Text to Speech and SEEING AI APIs. These services enable a read-aloud function with image descriptions as well aiding visually impaired users and others as well, such as those listening while driving. Next, learn about a multilingual chatbot powered by Azure OpenAI and Speech to Text capabilities. This inclusive tool breaks language barriers and assists users with mobility challenges, showcasing how accessible design benefits diverse audiences. Through this exploration, I hope to inspire you to consider the vast possibilities of accessible design, encouraging you to go beyond mere regulatory compliance and truly innovate for inclusivity.

Azure AI – Shaping an accessible world

Dive into the future of AI-driven development with Autogen Framework, an open-source tool that revolutionizes how we build complex workflows using conversational agents and patterns. In this demo-ful session, we’ll explore the fundamental concepts and conversational patterns essential for crafting multi-agent AI applications within the .NET ecosystem.

The session will try to help you to build a robust understanding of how to leverage large language models (LLMs) to orchestrate sophisticated, agentic workflows. Topic covered:

Core Concepts: Get to grips with the foundational elements of the Autogen Framework.

Conversational Patterns: Learn how to design and implement effective conversational agents that can collaborate seamlessly.

Practical Applications: Walk through real-world scenarios and use cases to understand the framework's potential.

Unleash the Power: Crafting Agentic AI .NET Apps with Autogen Framework

Discover how GitHub Copilot's Private Extensions are reshaping the development landscape by providing AI-powered, context-aware code suggestions tailored to your organization's unique needs. This session will explore how these extensions enhance productivity, streamline collaboration, and boost code quality by leveraging proprietary knowledge. Learn how to securely integrate Copilot into your workflows, transforming the way your teams code, refactor, and innovate. Perfect for developers looking to harness AI to optimize their custom projects.

How GitHub Copilot Private Extensions Transform the Coding Experience

Discover how to access Azure Open AI from Azure SQL, querying your data with natural language and opening a world of possibilities

Azure SQL and Open AI using T-SQL

Keeping the production environment under control is becoming increasingly challenging, but luckily, with AI-powered tools, you can get the assistance you need by implementing true AIOps.

In this session, we will explore and test the various tools Azure Monitor offers to automate multiple IT Operations tasks with AI support, from log analysis to alerts and workload optimization.

Identify and mitigate production issues by adopting AIOps in Azure Monitor

Join us to discover the transformative potential of Azure AI Speech across diverse industries. Azure AI Speech empowers businesses with advanced capabilities such as speech recognition, text-to-speech, and speech translation, facilitating the creation of voice-enabled applications that significantly boost operational efficiency.
We will kick off the session by introducing Azure AI Speech services and their wide-ranging applications, emphasizing key features like real-time speech recognition, text-to-speech conversion, and speech translation. You will learn how to set up and configure Azure AI Speech for real-time speech-to-text conversion, with practical use cases including dictation, call center assistance, and live meeting captioning.
The session will delve into text-to-speech capabilities, demonstrating how to convert text into natural-sounding speech using Azure AI, with applications in virtual assistants, accessibility tools, and automated announcements. Additionally, we will explore speech translation for multilingual interactions and its use cases in global support and cross-language communication.
We will outline the steps to integrate Azure AI Speech into existing applications and share best practices for deploying and scaling speech-enabled solutions. Through case studies and success stories, we will illustrate how businesses leverage Azure AI Speech to enhance operational efficiency and accessibility.  Unlock the potential of Azure AI Speech and transform your interactions with cutting-edge voice technology. Don't miss this opportunity to stay ahead in the AI-driven future—register now and take the first step towards revolutionizing your business!


Revolutionizing Interactions with Azure AI Speech

"Threat or Opportunity? — Enhancing Software Development Efficiency with AI" explores the transformative role AI tools, like ChatGPT, play in the software development process. As AI continues to evolve, it presents both challenges and possibilities for developers. This presentation will focus on how AI can enhance productivity, reduce repetitive tasks, streamline code reviews, and support decision-making, ultimately making development workflows more efficient. It will dive into practical examples of integrating AI in everyday tasks, helping developers to leverage these technologies as powerful allies rather than seeing them as a threat.

Threat or Opportunity? — Enhancing Software Development Efficiency with AI

Voice AI agents are transforming the operational landscape for small and medium-sized businesses (SMBs), offering automation and enhanced customer experiences. However, building these systems comes with unique challenges. This talk will explore the technical and operational hurdles in building and deploying voice AI agents tailored to SMB verticals. We will also delve into cutting-edge advancements, such as multi-agent system development, multimodal integration for richer interactions, and the use of reinforcement learning for continuous improvement. Lastly, the session will discuss evaluation frameworks—beyond traditional metrics—to measure performance, usability, and business impact in real-world settings.

Voice AI Agents for SMB Operations

Explore how to harness the power of large language models (LLMs), Cognitive Services, and intelligent tools to create innovative solutions for modern business challenges. Through real-world examples and a live demonstration, learn how to integrate Azure AI into your workflows, automate processes, and build AI-powered applications seamlessly. Whether you're a developer, data scientist, or tech enthusiast, this session will equip you with actionable insights and practical knowledge to accelerate your AI journey with Azure.

Innovation with Azure AI: From LLMs to Cognitive Services

In this session we will have an introductory overtime of Microsoft Fabric. Understanding the components and seeing examples of the different personas approach. Will review the approach for Data Factory, Data Engineering, Data Science, Data Warehousing, Real-Time Analytics and Power BI. Finally, we will see how to start getting involved with the new platform for Data analytics for the era of AI

Introduction to Microsoft Fabric - Data analytics for the era of AI

In this session, we will explore Phi Models, one of the most innovative approaches within Small Language Models (SLMs). We’ll discuss how these models are designed to deliver exceptional performance in natural language processing tasks while maintaining a compact structure, making them more efficient and accessible than traditional large-scale models. 

The session will cover how Phi Models leverage the strengths of SLMs to tackle challenges such as reducing computational resources, improving inference speed, and optimizing applications on resource-constrained devices. Additionally, we will look at their real-world applications and the impact they are having on the development of more agile and sustainable AI solutions.

Phi Models: Innovation in Small Language Models for efficient language processing

Join Alon Fliess, Azure AI Platform MVP and Microsoft Regional Director, for a transformative lecture exploring how Semantic Kernel is ushering in a new era of software development. In a world where deterministic algorithms are no longer sufficient on their own, developers and architects must embrace hybrid solutions that integrate deterministic logic, heuristic approaches, and cutting-edge AI technologies such as generative AI and machine learning.

This session dives into the paradigm shift in software engineering, highlighting the challenges and opportunities of building systems that must evaluate and manage non-deterministic results alongside deterministic logic. With the increasing use of AI agents that generate solutions and validation agents that safeguard and guide their output, responsible AI practices are becoming central to modern software development.

Attendees will gain insights into the capabilities of Semantic Kernel, a project that seamlessly integrates C# with AI-driven functionalities. By combining traditional programming approaches with AI capabilities, developers can craft systems that are not only intelligent but also adaptable to complex and evolving environments. This lecture will showcase real-world applications of Semantic Kernel, demonstrating its potential to revolutionize diagnostics, analysis, and decision-making in sophisticated systems.

By the end of the session, C# developers and architects will have a deep understanding of the concepts and methodologies driving this paradigm shift. They will leave equipped with the tools and knowledge to harness Semantic Kernel and its capabilities, bridging the gap between conventional programming and the dynamic, AI-powered future of software development.