Microsoft Reactor Bengaluru

This session demystifies AI agents in 2025, cutting through the hype to provide practical knowledge on building effective agents. Participants will explore essential use cases, understand the components of AI agents, and learn how to implement them using Microsoft's Azure AI Agent Service, Semantic Kernel, and Autogen frameworks. The session includes Live demonstrations to showcase real-world applications of agentic AI.

The session will focus on practical approaches to building AI agents using Microsoft's agentic services and frameworks, emphasizing understanding key patterns, components, and implementation techniques in Azure.

What will you learn from this session?
Attendees will learn how to identify appropriate use cases for AI agents and gain experience building agents using Azure AI Agent Service, equipping them with the skills to develop their own agentic solutions.

Pre-requisites:
https://aka.ms/MSAzAIFundamentals-AIOverview

This session will guide attendees through the process of building a Knowledge Store using Azure AI Search, enabling persistent storage of enriched data for downstream analytics and processing. Participants will explore the concept of projections, how to define a knowledge store within a search enrichment pipeline, and how to access and visualize the structured output. The session includes a hands-on exercise to implement a full knowledge store pipeline in Azure.

Read More - https://aka.ms/AzureAI-Search

The session will focus on:
Understanding projections and how to define them.
Defining and configuring a knowledge store.
Persisting enrichment outputs (e.g., key phrases, entities, image content) in a queryable form.
Accessing and analyzing the stored data using tools like Power BI or custom dashboards.
Hands-on Exercise: Build and test an enrichment pipeline with a connected knowledge store.

What will the attendees or a Start-up learn from the session?
Attendees, including start-ups, will gain the following:
How to extend AI Search capabilities with a knowledge store.
Techniques to extract, project, and store enriched data.
Real-world use cases for building intelligent knowledge apps.
How to integrate the knowledge store with reporting tools and downstream processing systems.
Best practices to design and scale knowledge stores in AI-driven applications. 🚀

Speaker BIO- Viswanatha Swamy
He is an aspirant Software Architect and currently, he works at Applied Information Sciences. He is passionate about C#, ASP.NET, Azure, Performance testing/tuning,.NET Core, and Docker. He loves to learn about new technologies.

Social Handle- https://twitter.com/vishipayyallore

Pre-requisites:
AI-900
Some experience in AI
Willing to learn AI

The Model Context Protocol (MCP) is rapidly emerging as the new open standard for connecting AI models—like large language models (LLMs)—to the external tools, data sources, and business systems that power modern workflows. In this session, we’ll demystify MCP and show how it transforms AI integration from a tangled web of custom connectors into a streamlined, scalable, and secure architecture—think “USB-C for AI.”
You’ll learn how MCP enables seamless, two-way communication between AI assistants and your organization’s data, whether it lives in Microsoft 365, Azure-based services, GitHub, Slack, or custom business applications. We’ll break down the core components of MCP—hosts, clients, and servers—and demonstrate how they work together to let AI applications retrieve data, invoke tools, and maintain rich context across platforms.

The session will focus on:
Through practical examples, we’ll explore:
•The fundamentals of MCP’s client-server architecture and how it standardizes AI-to-tool communication
•How to build or deploy MCP servers to expose your data and tools to AI, including with C# and .NET in Microsoft environments5
•Real-world use cases: from automating code reviews with Claude and GitHub to integrating AI with Microsoft’s productivity suite
•Security, authentication, and best practices for safe, enterprise-ready deployments47
•The growing ecosystem of MCP-compatible tools and how to future-proof your AI strategy.

What will the attendees learn from this session?
Whether you’re a developer, IT leader, or business innovator, you’ll leave this session with a clear roadmap for leveraging MCP to unlock the full power of AI in the Microsoft world—enabling smarter, more connected, and more productive solutions for your team or organization.

Read More - https://aka.ms/MS-Copilot-LearningHub

Event Host: Kamal Shree Soundirapandian
Kamal Shree is a Developer Advocate at Microsoft. She is a Google Developer Expert, YouTuber (Whatsupcoders) with 12 years of experience in Web Technologies, Android, Flutter, and HarmonyOS. She has worked for multinational firms in India, Netherlands, and the USA. She is also a Mentor and Open-Source Community Builder.

Social Handles of the speaker:
https://twitter.com/whatsupcoders
https://www.linkedin.com/in/kamalshree/

Speaker Bio - Alistair Pugin
Alistair Pugin has worked in various capacities in multiple verticals from retail – manufacturing to government, spanning 50 to 50000 users utilizing all aspects of pure Enterprise Information Management.
His current role as a Microsoft Security and M365 Apps and Services MVP in South Africa, provides him with a mechanism to assist companies with designing their Cloud collaboration environment using “best of breed”, proven methodologies to foster innovation and growth through a Content Services framework while incorporating a Knowledge Management strategy.
It is with this vision that he strives to provide customers with a platform for Business Productivity Enrichment, establishing communities of practice through proper analysis and building a knowledge economy that is both quantifiable and achievable.

Social Handle of the speaker-
@alistairpugin everywhere. (X, IG, Threads, BlueSky, LI)

Pre-requisites:
A base level understanding of Microsoft 365 Copilot, Copilot Studio

Architecting AI Safety: Explore Azure AI's technical guardrails implementing trustworthy AI principles. We'll deep-dive into Content Safety implementation, model control mechanisms, configurable content filtering, and blocklist deployment. Learn practical approaches to responsible AI development through Azure AI Services' system messages and limited access features—essential knowledge for production-ready AI systems.

The session will focus on:
Practical implementation of Azure AI safety guardrails, including content filtering, blocklists, and system message controls for responsible AI development.

What will the attendees learn from this session?
Attendees will master Azure AI safety implementation techniques, including content filtering configuration, blocklist deployment, and system message controls for building trustworthy AI applications.

Read More:
https://aka.ms/ResponsibleAIatMicrosoft
https://aka.ms/AIforAzureAIFoundry

Speaker Bio- Sunil Kumar
Cloud Consultant and corporate trainer in cloud technologies. 16+ years of total working experience in teaching, training, and industry. An aspiring community builder, content creator on YouTube, Medium, and LinkedIn. A technical events and conferences enthusiast and toastmaster.

Social Handles of the speaker:
https://www.linkedin.com/in/cloudwithsunil/
https://www.youtube.com/@sunilkumarOnCloud
https://medium.com/@suniel.vns

Pre-requisites:
Azure subscription for the hands-on, Azure AI Foundry basic knowledge