Search, Summarize, Simplify: Copilot Multi-Agents for Everyday Productivity

In a world overflowing with emails and documents, staying productive means finding the right information quickly and acting on it without delay.

In this session, discover how to build and deploy Copilot Multi-Agents that simplify everyday work by automatically summarizing emails, extracting action points, and searching SharePoint for related content. We will explore how AI, Microsoft Copilot, and Power Platform come together to create intelligent workflows that save time, reduce context-switching, and empower users to focus on what really matters.

Whether you’re an IT professional, a developer, or a business leader, you will get to see and experience a live demo showing how AI agents can become your everyday productivity partners.

Transcript

1. Search, Summarize, Simplify: Copilot Multi-Agents for Everyday Productivity
2. Agenda • Understanding the Problem Landscape • How SharePoint Search works ? • Key challenges and limitations • Introduction to Copilot Agents • Understanding the RAG model • Integrating Copilot Agents with RAG and SharePoint • Demo • Q/A
3.  Failed to realize that non-business consumers would drive the market.  Poor reaction to the new app economy that made Apple and Android devices become popular.  Didn’t foresee the mobile phone market would evolve from a simple communication device to a full-mobile experience entertainment device.  The executives remained on the QWERTY keyboard in the phase of touch screen devices. Agents are coming.. Similar cases : Nokia, Yahoo, Hotmail, Sun Systems, Kodak.
4. Use Case “We store all our data and contract documents in SharePoint sites. These contracts are with various vendors and clients, and they come in multiple formats -Word, Excel, PowerPoint, PDF, ZIP files, and even documents with embedded content. The requirement was to make it easier to search within these documents -for example, to quickly find who signed a contract, what terms and conditions were agreed upon, or what discussions took place. Users were facing several challenges with the standard SharePoint Search. To address this, we developed the “Treaty Search” AI-powered Power Apps solution, built with Copilot Agents and the RAG (Retrieval-Augmented Generation) model on SharePoint sites. This solution makes document search faster, smarter, and more intuitive, helping users quickly find the right information with better accuracy and context..”
5. We spends 20–30% of their time searching for information.. • Time wasted in finding the right file. • Large repositories, unstructured data (PDF, Word, Excel, Images). • Metadata mismatch (wrong tags, inconsistent naming). • Content buried deep in documents such as embedded document search. • File format limitations such as excel, zip, embedded doc Key Challenges
6. Improvements over the years .. Reference : https://evolvous.com/ai-powered-search-in-sharepoint 2023 • Modern Search Experience Transformation • MS Search was introduced, where AI and MS Graph were used together for personalizing search results. 2024 • Highly Intuitive Search Process • Recognizing synonyms and similar concepts and identifying and aligning with user intent to improve the quality of searches began to make way. 2025 • PnP Modern Search Open-Source Introduced • Customizable web parts can help you opt for advanced filtering and use dynamic search experiences.
7. Search Options  SharePoint search: Originally built for the on-premises SharePoint, it is known for its tight integration with document-level user permissions for improving and making access control stricter.  Microsoft Search: SharePoint currently uses machine learning. It has a limited set of customization options and lacks document-level security settings.  Azure AI Search: It comes with sophisticated AI search capabilities. It would need suitable API implementations. Security trimming could need some work and does not add native document level security settings. All of these take you one step closer to finding the right file in a short amount of time.
8. How Search works ? Reference : https://learn.microsoft.com/en-us/sharepoint/dev/general-development/search-in-sharepoint  Crawl component takes care of crawling the content sources such as (SharePoint sites, websites & file shares etc…) and extracts the crawled properties and metadata and sends that to the content processing component.  Content processing component receives the information from the crawl component and then processes and sends it to the indexing component. It also interacts with the analytics processing component and is responsible for mapping crawled properties to the managed properties.  Analytics processing component takes care of analyzing what users are querying on and how they interact with the results. This information is used to determine relevance, generate recommendations and also used for generating search reports.  Index component takes care of receiving the information from the content processing component and writes it to the search index. It also takes care of handling the queries and sends back the results to the Query processing component.  Query processing component handles incoming query requests and sends them to the indexing component for results. It also takes care of query optimization.
9. Limitations with Sharepoint Search • Indexing Gaps – SharePoint skips or incompletely indexes some formats (ZIPs, embedded docs, scanned PDFs), and limits full-text indexing to the first few MB, so large or complex files may be partially searchable only. • Metadata Mapping Issues – Custom fields aren’t searchable until manually mapped to managed properties; unmapped metadata also won’t appear in filters or refiners. • Weak Relevance Ranking – Keyword-based search often misranks results; lacks semantic understanding (e.g., “lawyer” ≠ “attorney”) and undervalues document context. • Performance & Freshness Delays – Crawls can lag minutes → hours; misconfigurations cause stale indexes where updated or new files don’t appear. • User Experience Friction – Search syntax feels technical, irrelevant hits overwhelm users, and “Classic vs Modern” search trade-offs (customizability vs UX) confuse teams. • Security & Advanced Query Limits – Permission-based trimming hides valid files, while search lacks flexible wildcards and intuitive multi-condition filters.
10. What are agents? Agents are apps that use AI to reason, plan, connect to systems and complete tasks working alongside or on behalf of a person, team or organization Simple Advanced Retrieval Task Autonomous Agents vary complexity and capabilities depending on your need
11. A range of tools for agent creation No code Pro code For end users Agent builder For makers Copilot Studio For developers Copilot Studio, Azure AI Foundry & Visual Studio Data protection, agent sharing & usage limits, and reporting & cost management
12. RAG (Retrieval Augmented Generation) Reference : https://www.k2view.com/blog/rag-architecture/#RAG-Architecture-Step-by-Step How it works ? You ask: “Show me the latest treaty where ABC Insurance agreed to renewal terms.” Without RAG: SharePoint search might show a long list of files, and you have to open each one. With RAG + Copilot Agent: You get a concise, AI-generated summary: “The renewal treaty signed on 12 March 2024 with ABC Insurance includes a 10% rate adjustment and was approved by John Smith.”
13. Vector(Semantic) Indexing It’s a process where each word is given a number that represents its meaning, and similar words are grouped together. In keyword-based search, the system only looks for exact words you type. But in vector (semantic) search, it also finds words with similar meanings — even if they’re not exactly the same.
14. KEYWORD VS SEMANTIC UNDERSTANDING • Keyword Questions • “What is the policy on annual leave?” • “Do we have a policy regarding workplace safety?” • “How many sick days are employees entitled to annually?” • “What are the guidelines for filing a harassment complaint?” • Semantic Questions • “How can an employee request a leave of absence for mental health reasons?” • “What steps should be taken if an employee feels unsafe at work?” • “Can part-time employees also participate in the health benefits program?” • “What should a manager do if they observe discriminatory behavior?” Specific terms found in the documents More nuanced understanding required
15. Prompts
18. Let’s Build Agents (Demo)
19. Gandalf Challenge https://tinyurl.com/ainov2025
20. How to select LLMs • The choice of model determines the intelligence and capabilities of your agent. • Choose based on: • Complexity of tasks • Access to tools/plugins • Performance vs. cost Reference : https://platform.openai.com/docs/models https://platform.openai.com/docs/pricing
21. Tips • Understand file types, limitations & behaviors. • Get at-least 1 Microsoft365 license in your tenant. • Use the Generative AI Orchestration Feature. • For SharePoint sources, if you don't have a Microsoft 365 Copilot license in the same tenant as your agent, generative answers can only process files up to 7 MB in size. You must also turn off the Enhanced search results feature. • If you have a Microsoft 365 Copilot license in the same tenant as your agent, the maximum file size is 200 MB. You must also turn on the Enhanced search results feature. • Larger files can be stored in SharePoint and are returned by a Microsoft Graph search, but aren't processed by generative answers. As an alternative, you can upload your own files, which can be up to 512 MB in size.
22. Summary Discussed the problem statement and current limitations within SharePoint Search. Gained a understanding of how SharePoint Search works and its core mechanisms. Identified the key challenges and issues impacting search accuracy and relevance. Explored potential approaches to build intelligent Copilot Agents to enhance search capabilities. Discussed the RAG model and how it can be applied. Evaluated how Copilot Agents combined with the RAG model can help overcome existing SharePoint Search limitations and deliver AI-augmented insights.
23. Q/A
24. Thank you..