🚀 Exciting Announcement: Join Us for AI DAYS Tunisia – December 2025! 🤖
Get ready for two incredible days of innovation, inspiration, and interaction at AI DAYS Tunisia — a free, community-powered event celebrating the full spectrum of Artificial Intelligence!
📍 Venue: Blue Marine Hotel Hammamet
📍 Address: Yasmine Hammamet 8050, Tunisia
📅 Day 1 – December 19, 2025 | 🕒 9:00 AM CET
📅 Day 2 – December 20, 2025 | 🕒 9:00 AM CET
💸 Admission: 100% Free – Registration Required
🌟 Why You Shouldn’t Miss This:
✅ AI Model Development – Learn to build and fine-tune models using Microsoft AI, OpenAI, Hugging Face, and more.
✅ Real-World Applications – Discover how AI is transforming healthcare, education, finance, and climate tech.
✅ Startup Showcases – Meet Tunisia’s most promising AI startups and explore their groundbreaking solutions.
✅ Live Demos &amp; Q&amp;A – Watch AI in action and engage directly with experts.
✅ Healthcare &amp; AI – Explore how AI is revolutionizing diagnostics, treatment, and patient care.
🎤 Event Highlights:
💡 Inspiring Keynotes – Hear from global AI leaders, including Microsoft AI experts, sharing their vision for the future.
🧠 Thought-Provoking Panels – Dive into discussions on AI ethics, diversity, regulation, and societal impact.
🛠️ Hands-On Workshops – Gain practical skills with tools like Azure AI, Copilot, and open-source frameworks.
🤝 Networking Opportunities – Connect with developers, researchers, entrepreneurs, and decision-makers.
🎉 Interactive Exhibits &amp; Entertainment – Enjoy AI-powered art, games, and immersive experiences for all ages.
💼 Want to Sponsor or Participate?
We welcome sponsors, partners, and community contributors! Reach out to us at :&nbsp;Jamel Abed: &nbsp;<a href="mailto:jamel.abed@gmail.com" target="_blank">jamel.abed@gmail.com</a>&nbsp;Dr. Mohamed Touati:&nbsp;<a href="mailto:mohamed.touati.enis@gmail.com" target="_blank">mohamed.touati.enis@gmail.com</a>

🛏️ Need Accommodation?
We’ve secured a special group rate at the Blue Marine Hotel Hammamet our event venue. For hotel bookings, transfers, or other assistance, email us directly.
🎫 Reserve your FREE spot today and be part of Tunisia’s most exciting AI event of the year!
Stay tuned for more updates, speaker announcements, and surprises. We can't wait to welcome you!

AI DAYS TUNISIA

This session explores how modern OCR/NLP pipelines can transform document processing across multiple sectors, with methodologies applicable to:

1. Healthcare Use Cases:
*Patient record digitization
*Insurance claim processing
*Clinical trial data extraction

2. Financial Services:
*Invoice processing
*KYC document verification
*Contract analysis

3. Government & Education:
*Administrative form processing
*Academic record management
*Archive digitization

"Cross-Industry Document Intelligence: From Healthcare to Enterprise Applications"

Description: This session dives into PixOCR, a SaaS solution revolutionizing patient registration by automating data extraction from IDs, insurance cards, and referral letters to populate EHRs. Attendees will learn:  
  
1. Technical Architecture:
OCR Stack: Hybrid approach using Tesseract (for structured fields) and Google Cloud Vision (handwritten text).  
 NLP Pipeline: spaCy for general entity recognition + BioClinicalBERT (fine-tuned on MIMIC-III) for medical context understanding.  
Integration: Apache NiFi for data ingestion, FastAPI backend, and React frontend.  
  
2. Compliance & Real-World Impact: 
- GDPR/HIPAA: Built-in anonymization and audit trails for sensitive data.  
- Case Study: Reduced patient check-in time from 15 to 5 minutes at a Tunisian clinic.  
 
3. End-to-End Workflow Demo:  
- Document upload → OCR → EHR integration (via FHIR APIs with systems like Epic) → Compliance checks.  


"PixOCR in Healthcare: AI-Powered Patient Data Extraction & Registration"

Have a question? We may have the answer. Or not. You won't know until you ask. 
Disclaimer: Food not provided, but on request we may recommend recipes for amazing dishes.

AAAAA - Ask Anyone Anything About Anything

What happens when GitHub Copilot’s AI-powered coding collides with Azure AI Foundry’s agent-based architecture? You get a developer-centric workflow that’s fast, flexible, and dangerously productive.

This session walks through building an AI-powered support ticket routing system using GitHub Copilot and Azure AI Foundry. You’ll see how Copilot helps scaffold the agent logic, wire up external APIs, and generate smart responses—all inside your editor. Then we’ll connect it in Azure AI Foundry to run the logic as an intelligent, stateful agent that classifies, routes, and responds to support tickets in real-time.

A short live demo will show how to:

- Use GitHub Copilot to rapidly scaffold code for an AI agent
- Connect the agent to external APIs like Microsoft Graph or Teams
- Deploy the agent in Azure AI Foundry with minimal setup

Whether you're exploring multi-agent systems or just want to see Copilot used beyond boilerplate generation, this session gives you practical tools and a real-world use case to build on.

Agents, Code, and Chaos: GitHub Copilot Meets Azure AI Foundry

The talk is about highlighting the different ethics related to the use of AI.

AI Ethics

Corporates and startups share the same ultimate goal: delivering ROI from AI. Yet, their paths are radically different. Drawing on experience with leading financial institutions (Allianz, AXA, SCOR, CNP) and successful ventures (Memorality, SnapScout & Makassur "launching in October 2025"), Mariem reveals what truly works to scale AI in finance/ insurance — from governance that accelerates rather than slows, to robust data & MLOps pipelines, compliance built-in from day one, and product design that drives adoption. 

Attendees will walk away with actionable frameworks, proven metrics, and practical checklists to move AI from POC to profit — at scale, securely, and in full compliance.

AI in Finance: From POC to Profit — Bridging the Corporate × Startup Gap

Discover how Generative AI can handle your repetitive tasks, boost your creativity, and help you work smarter not harder. In this 45-minute session, you’ll learn practical prompting techniques, see real-world examples, and watch how AI that turn rough ideas into polished results in seconds. Leave ready to make AI your most productive teammate.

AI, Please Do My Job (…But Better)

Refactoring is a critical task in software maintenance and is usually performed to enforce better design and coding practices, while coping with design defects. The Extract Method refactoring is widely used to merge duplicate code fragments into a new method. Several studies attempted to recommend Extract Method refactoring opportunities using different techniques, including program slicing, program dependency graph analysis, change history analysis, structural similarity, and feature extraction. However, irrespective of the method, most existing approaches interfere with the developer’s workflow: they require the developer to stop coding, analyze the suggested opportunities, and consider all refactoring suggestions in the entire project without focusing on the development context. To increase the adoption of the Extract Method refactoring, in this tutorial, we aim to show the effectiveness of machine learning and deep learning algorithms for its recommendation while maintaining the developer's workflow. Finally, we demonstrate a case study on how LLMs can be useful to recommend the Extract Method refactoring.

AI-Driven Software Engineering and Quality Assurance

This workshop will introduce attendees to many of AI and Machine Learning services and tools in the Microsoft world, while providing for hands-on experience. 

This is a beginner level workshop into AI/ML (no pre-requisite of AI/ML knowledge, but knowledge of programming is expected).

Machine learning is a data science technique that allows computers to use existing data to forecast future behaviour, outcomes, and trends. By using machine learning, computers learn without being explicitly programmed. Forecasts or predictions from machine learning can make apps and devices smarter. For example, when you shop online, machine learning helps recommend other products you might want based on what you've bought. When your credit card is swiped, machine learning compares the transaction to a database of transactions and helps detect fraud. And when your robot vacuum cleaner vacuums a room, machine learning helps it decide whether the job is done.

Microsoft Fabric, Azure Synapse Analytics, Azure Databricks (and generically Spark, using Python, R, Scala or SparkSQL), and Azure Machine Learning,  are cloud services that you can use to train, deploy, automate, and manage machine learning models, as well as process ETL operations on your data, all at the broad scale that the cloud provides.

Azure AI Services (formerly Cognitive Services) are APIs/SDKs/services available to help developers build intelligent applications without the need for AI or data science skills/knowledge. Azure AI Services enable developers to easily add AI features such as emotion and video detection; facial, speech, and vision recognition; and speech and language understanding – into their applications. The goal of Azure AI Services is to help developers create applications that can see, hear, speak, understand, and even begin to reason.

Azure AI Foundry Agents/Assistants, Microsoft 365 Copilot Studio (formerly Power Virtual Agents) and Azure Bots use these services integrated with the Bot Framework to provide a conversational experience that tries to resemble human interaction. Bots can be deployed in channels such as Teams, Slack, Skype, Facebook, Alexa, IoT, or any other you can imagine.

Attendees that want to try examples during the workshop need to have:
• Laptop;
• Have an Azure subscription (e.g. set up a free trial at https://azure.com/free , use an Azure Pass free trial, or use a paid subscription);
• For Microsoft Fabric, have an active Microsoft 365 subscription with the possibility of starting a free Fabric Trial;
• Your preferred development IDE (e.g. the free Visual Studio 2022 Community Edition or Visual Studio Code) already installed, to access online webservices;
Note: For AI services and APIs, C# and Python examples will be covered, but equivalent examples exist in other programming languages.

Contents:
1. AI and Machine Learning basics;
2. Introduction to Azure AI Foundry (covering LLMs like GPT models and other Model-As-A-Service options);
3. Introduction to other Azure AI Services;
4. Introduction to Agents using Copilot Studio, Agents and Assistants using AI Foundry, integration with Microsoft 365 Copilot, etc.;
5. Introduction to Machine Learning and ETL technologies: Microsoft Fabric, Azure Synapse Analytics, Azure Databricks, Azure Machine Learning, etc.;
6. Architecting AI technologies in real life;
7. Q&A.

#Azure #AzureOpenAI #AzureAIFoundry #MicrosoftFabric #AzureML #MachineLearning #SynapseAnalytics #Databricks #CognitiveServices #AzureBot #CopilotStudio #BotFramework

Azure AI Workshop [V6]

AI adoption is accelerating across industries, but implementation alone doesn’t guarantee success. Organizations often focus on surface-level engagement metrics—system logins, automation rates, and AI-driven outputs—without assessing whether employees truly understand and leverage AI capabilities. 

The result? Billions lost in unrealized ROI as AI tools remain underutilized or misapplied.

This session redefines AI adoption by introducing a proficiency-first framework that ensures organizations extract real value from their AI investments. 

Attendees will gain insights into:
•	The hidden gap between AI implementation and actual workforce adoption
•	Why traditional tech adoption metrics fail to capture AI utilization
•	Practical strategies for measuring AI proficiency beyond engagement statistics
•	The impact of AI-driven decisions on operational efficiency, revenue, and customer experience

Beyond Automation: Measuring AI Proficiency to Unlock True Business Value

AI is evolving — and it’s time to think beyond simple chatbots.

In this session, learn how to design Autonomous AI Agents using Microsoft Copilot Studio that not only respond to user queries, but also proactively complete tasks, orchestrate workflows, and act with context and purpose.

We’ll explore how Copilot Studio now supports multi-step reasoning, adaptive planning, and integrations across Microsoft 365 and external APIs — unlocking a new generation of agents that go beyond conversations to drive real action.

Topics we’ll cover:

What makes an AI Agent autonomous (vs a traditional chatbot)

Building multi-step reasoning and action planning into your Copilot workflows

Integrating Copilot Agents with Power Platform, Dataverse, and external systems

Real-world examples of Autonomous Agents in enterprises and government

Whether you're looking to automate internal processes, client services, or complex decision-making, this session will help you move from reactive chatbots to proactive AI agents.

Beyond Chatbots: Create Autonomous AI Agents Using Microsoft Copilot Studio

In today’s global workplace, delivering content in multiple languages is no longer a luxury—it’s a necessity. 

You’ll learn how to go beyond out-of-the-box multilingual features in SharePoint and leverage the power of Power Automate and Azure Translator to automatically translate site pages, ensuring consistent communication across languages.

Whether you're managing a corporate intranet or a project site, this session will show you how to make content accessible for everyone—no matter what language they speak.

Breaking the language barrier: Multilingual SharePoint with Azure Translation Services

Dive into the world of intelligent automation in this interactive session on creating autonomous agents using Copilot Studios. Whether you're a beginner or an experienced developer, this workshop will guide you step-by-step through the process of designing, building, and deploying autonomous agents tailored to solve real-world challenges. Learn how to harness the power of AI tools, integrate decision-making capabilities, and streamline workflows with ease. By the end of the session, you'll have the practical knowledge to implement autonomous agents.

Building Autonomous Agents with Copilot Studios

Applications of Remote Sensing and Geographic Information Systems in Water Resources Management

Coastal Flood Risk Forecasting in Libya by 2100 using Google Earth Engine”

En tant que développeur, j'attendais que l'IA m'aide dans mon développement. Mais en faisant ça, je suis passé a côté du comment l'IA pouvez m'aider. Et si on ne devait pas plutôt utiliser l'IA sur les autres tâches pour se concentrer essentiellement sur le développement/réalisation. Nous verrons comment utiliser l'IA dans la vie d'un projet Power Platform, depuis l'expression de besoin, en passant par la réalisation jusqu'au livrable de la documentation.

Comment l'IA est devenu mon assistant de mes projets Power Platform

Copilot Notebooks are the AI-native workspace in Microsoft 365 — a single hub where your files, links, chats, and notes live side-by-side with Copilot’s reasoning. Notebooks transform scattered information into a living canvas where people and AI work together: drafting plans, exploring ideas, and turning insights into shareable, contextual outcomes.
In this session, you’ll learn the fundamentals of AI-Native work, when and how to use Copilot Notebooks, co-creation with both human and AI collaborators, and how to iterate and share results across Microsoft 365. You’ll also learn about the latest updates that make Copilot Notebooks a practical, everyday tool for the AI-First workplace—far more than just a place to store notes and chats.

What You’ll Learn:
- AI-Native Work Fundamentals: Learn how to adopt an AI-First mindset in your daily work: how to co-create with AI as a thinking partner, iterate in context, and move from one-off automation to continuous, human–AI collaboration.
- Mastering Copilot Notebooks: Learn when to choose Notebooks over documents or chat prompts, and how to use them for focused, contextual, AI-assisted collaboration in real-world scenarios.
- Human AI Iteration: See how people and AI can co-edit and evolve content inside a Notebook using Copilot to build, refine, and expand ideas together in real time.
- Administration Glimpse: Get a high-level look at what Copilot Notebooks are under the hood: where they’re stored, how access and permissions work, and what tools IT has for administration, governance and compliance.
Walk away from this session with moving from “I’ve heard of AI-Native work” to knowing practically how to bring AI into your day-to-day collaboration with Copilot Notebooks.


Copilot Notebooks: AI-native co-creating from zero to hero

In an AI-driven world, fragmented data silos limit the potential of intelligent insights and automation. This session explores proven strategies to harmonize enterprise data across systems, creating a unified, consistent foundation ready for advanced analytics, machine learning, and AI deployments.

We will cover key approaches for aligning disparate data sources, enhancing data quality, and implementing scalable data models using the Microsoft Data Platform and AI tools.

Attendees will learn how data harmonization accelerates AI adoption, improves decision-making, and unlocks new levels of operational efficiency across industries.

Data Harmonization Strategies for AI-Ready Enterprises

As Copilot Accelerators at CopilotFactory, we specialize in transforming business challenges into intelligent, scalable solutions using Microsoft Copilot technologies. This session offers a behind-the-scenes look at our structured delivery model—from initial assessment and ideation to solution design, development, and deployment.

Learn how we accelerate Copilot adoption through tailored strategies, modular frameworks, and real-world use cases that drive measurable outcomes. Whether you're a business leader, developer, or transformation strategist, this session will provide actionable insights into how CopilotFactory delivers value at speed and scale.

Delivering Intelligence: The CopilotFactory Approach from Assessment to Action

Disabilities are not limitations. Technology helps balance the scale. 
This session shows some practical examples for real world usage, and you are encouraged to use these technologies to make your projects even more amazing and welcoming to everyone.

Develop for Inclusion using AI: An Azure Story

Rejoignez-moi pour une journée immersive et interactive où vous apprendrez à concevoir vos propres agents personnalisés avec Microsoft Copilot Studio !

Durant cet atelier, vous allez :

• Créer des agents sur mesure : apprenez à bâtir facilement un agent conversationnel qui répond à vos besoins métiers.

• Ajouter de l’intelligence et de l’action : connectez votre agent à vos sources de données, intégrez des actions via Power Automate et personnalisez ses réponses avec de la connaissance métier.

• Déployer en toute simplicité : découvrez comment publier et gérer vos agents à l’échelle de votre organisation, en toute sécurité.

Que vous soyez débutant curieux ou expert de la Power Platform / Microsoft 365, cet atelier est pensé pour vous donner les clés, les bonnes pratiques et surtout l’envie de créer vos propres agents.

Développez vos propres agents avec Microsoft Copilot Studio

The NVIDIA Certified “Fundamentals of Deep Learning” session is an official hands-on training program designed and validated by the NVIDIA Deep Learning Institute (DLI). It provides participants with practical experience in building, training, and deploying deep neural networks using industry-standard frameworks such as TensorFlow and PyTorch.

This session is taught by an NVIDIA DLI Certified Instructor, ensuring a high-quality, interactive learning environment that blends theory with real-world applications.

Fundamentals of Deep Learning – NVIDIA DLI Certified

La conférence introduit le contexte global de l’IA générative et son impact sur les modèles économiques, avant de se concentrer sur ses applications concrètes dans la distribution pharmaceutique. Elle aborde les leviers d’optimisation qu’elle offre, les risques à anticiper, et propose une approche progressive fondée sur des projets pilotes, la formation des équipes et des partenariats ciblés.

Generative AI from the CEO’s perspective

Attending your session with strong motivation as an AI enthusiast driven by a sharp eye for innovation and a hunger for impact. Ready to absorb, challenge, and contribute to the future of intelligent systems.

Graph-Enhanced Retrieval: Scaling RAG with Temporal Intelligence

Use of AI is transforming healthcare at an unprecedented pace from diagnosis to personalized treatments. But with great power comes great responsibility. This keynote explores the ethical challenges and opportunities of medical AI, addressing issues of bias, transparency, patient trust, and accountability. Together, we’ll reflect on how to harness innovation without losing sight of humanity.

Healing with Code: The Ethics of AI in Healthcare.

Have you ever had a great idea for a business, but didn’t know where to start or felt overwhelmed by all the steps needed to make it real? You’re not alone. Most of us know that frustration. In this session, I’ll share my personal journey building StarterPilot, an AI-powered platform designed to help anyone turn their startup idea into a real project faster than ever.

We’ll walk through how AI is changing the game for entrepreneurs. Things like market research, finding the right name and logo, building a landing page, or launching your first website can finally be simple and even fun, even if you don’t have a technical background. I’ll do a live demo, showing how StarterPilot can take an idea from just a thought in your head all the way to a public launch that’s ready for real users. You’ll see how the platform provides instant feedback, branding assets, live landing pages, and even a way to show off your idea to the global community.

Special Bonus:
Because I believe in helping new creators and supporting our community, everyone attending this session will get free access to StarterPilot to try it out for themselves.

Let’s break down the barriers to entrepreneurship and see how AI can help more Tunisians (and beyond) turn their ideas into something real. No fancy jargon or tech background needed. Just bring your curiosity!

How AI is Empowering the Next Generation of Entrepreneurs

This provocative conference session title playfully merges the espionage mystique of James Bond with the cutting-edge world of AI agents, drawing inspiration from the “license to kill” concept while repositioning it as a “license to build” innovative 007 agents - such as self service, resumee or generative agents.

Join us for an exciting deep dive into the world of Microsoft Agents, where we’ll explore the cutting-edge capabilities of Copilot Agents, SharePoint Agents, Custom Agents, and Autonomous Agents. Whether you're a beginner or an experienced developer, this session will equip you with the tools and knowledge to create smarter, more efficient agents for your business needs.

We’ll break down the unique features of each agent type, demonstrate best practices with real-world business cases, and show you how to build them step-by-step. By the end of this session, you'll be ready to harness the power of these innovative agents to drive productivity and streamline your workflows.

Key Takeaways:
	• Discover the differences between Copilot Agents, SharePoint Agents, Custom Agents, and Autonomous Agents.
	• Gain insights into which agent to use for specific business needs.
	• Learn how to build and deploy these agents for maximum impact.

Don’t miss this chance to unlock the full potential of Microsoft Agents and take your automation skills to the next level!

License to Build: Copilot 007 Agents on Her Majesty’s Digital Service

Things are getting hot and heavy for AI, now that Azure AI Foundry is doing its sinister business. Forging devilish bots, torturing text and extracting diabolical knowledge, generating demonic cat images, or simply using someone else's not-so-malicious models (and then also the zero-humped Llama from a certain malevolent social network), and so much more villainy...
Disclaimer: Doctor Evil will now control your life and
║ The generated content was filtered due to triggering
║ Azure OpenAI Service’s content filtering system.
║ Reason: This response contains content flagged as Violence (very high).

Melting AI Foundry: An Evil Azure Story

Dans un monde où l’intelligence artificielle transforme profondément les modèles d’affaires et les expériences utilisateurs, la certification Microsoft Azure AI Engineer Associate (AI-102) se positionne comme une reconnaissance stratégique pour les professionnels IT et data. Cette session de 45 minutes vous plongera dans les compétences clés couvertes par cette certification : conception et mise en œuvre de solutions d’IA conversationnelle, intégration des services cognitifs, gestion de pipelines d’IA et déploiement dans des environnements sécurisés et scalables.

Nous explorerons ensemble la valeur ajoutée professionnelle de cette certification, aussi bien pour votre évolution de carrière que pour la compétitivité de votre organisation : amélioration des capacités de traitement du langage naturel, création de chatbots intelligents, automatisation de tâches complexes, et optimisation des processus métier grâce aux outils Azure Cognitive Services, Language Understanding et Azure Bot Framework.

Enfin, nous détaillerons les opportunités concrètes offertes par cette certification : accès à des projets à forte valeur ajoutée, reconnaissance sur le marché international, et contribution directe à la transformation digitale. Vous repartirez avec une vision claire des compétences à maîtriser, des cas d’usage réels, et des étapes pour réussir votre certification AI-102, le tout dans une perspective pratique et orientée résultats.

Microsoft AI Engineer AI-102

Learn how to build secure, enterprise-grade AI chatbots in Microsoft Copilot Studio—while ensuring your data stays compliant, discoverable, and trusted.

In this session, we’ll walk through how to harness Microsoft Copilot Studio to create conversational AI agents connected to internal data sources like SharePoint, Dataverse, and Microsoft 365. But building chatbots is just half the story.

We’ll also explore how to use Microsoft Purview to establish guardrails—ensuring data classification, access control, and compliance are built into your AI workflows from day one.

Key takeaways:

Connecting Copilot Studio to enterprise data sources

Designing user-friendly bot flows

Implementing data governance with Microsoft Purview

Building trust and control into AI conversations

Real-world examples of public sector and enterprise deployments

Responsible AI Chatbots: Building with Copilot Studio and Microsoft Purview Guardrails

Large Language Models (LLMs) hold transformative promise, but their scale makes traditional fine-tuning costly and impractical. This session explores Parameter-Efficient Fine-Tuning (PEFT) with a focus on LoRA (Low-Rank Adaptation) and QLoRA (Quantized LoRA), two breakthrough techniques that slash compute and memory requirements while enabling powerful customization. Attendees, from AI students and engineers to seasoned researchers, industry leaders, and startup founders, will gain a clear understanding of LoRA’s mathematical foundation, QLoRA’s advances, and real-world applications spanning Retrieval-Augmented Generation (RAG) and Multi-Task Learning (MTL). We’ll also discuss industry challenges, startup opportunities, and emerging research questions, showing how PEFT is lowering the barriers to LLM adoption, enabling scalable, cost-effective deployment across both research and enterprise contexts, and reshaping the trajectory of AI innovation.

Key Takeaways:

•	Grasp the core concepts and mathematical intuition of LoRA and QLoRA.
•	Understand the drastic efficiency gains over traditional fine-tuning.
•	Explore practical applications in RAG, Multi-Task Learning, with real-world use cases.
•	Identify current challenges and research opportunities driving future research directions in parameter-efficient LLM adaptation, including the balance between generalization and specialization.


Scalable LLM Fine-Tuning with LoRA & QLoRA: From Research to Entreprise Applications

This session explores key principles and practices for building ethically aligned, secure AI solutions. Covering responsible AI, generative model risks, identity management, agentic workflows, and governance strategies — it offers a vendor-neutral framework for developers and architects navigating the intersection of AI and cybersecurity.

Secure and Responsible AI Development

This session is tailored for business users, developers, and administrators, including PowerApps, SharePoint, and Azure developers.

It will delve into how to choose the most effective storage solution for your development project, whether you're creating in-house applications or purchasing from ISVs or Microsoft.

A detailed checklist will be provided to guide you before starting your project, ensuring the selection of the most appropriate storage option for your needs. The solutions discussed will include SharePoint Embedded, SharePoint Document Library, and Blob Storage. Each solution will be evaluated based on various requirements, with an in-depth examination of working with SharePoint Embedded.

The session will emphasize the significance of data for your company and projects, highlighting its role in informed decision-making, problem-solving, and process improvement. Security and compliance aspects of each storage solution will also be covered, ensuring your data is protected and meets regulatory standards.

Furthermore, the cost implications of each storage solution will be analyzed, aiding you in making budget-conscious decisions. The session will also explore how AI and Copilot agents can enhance your storage strategy, from automating data management to improving data analytics.

By the end of this session, you will have a clear understanding of the optimal storage solution for your needs.

Selecting AI-Optimized Storage Solutions for Modern Cloud Applications

In this demo-centric session you see how to create an example support desk agent team scenario with a no-code, multi-agent solution built with Azure AI Foundry Agent Service. 
We use a team of specialist agents—one grounded in documentation, one that gauges sentiment, and one that sends email —then wire them together under a single orchestrator agent and demo that in the Azure AI Foundry Agent Playground. You’ll watch how these agents work together: the knowledge agent tries to discover the answer, the sentiment agent measures frustration, the orchestrator resolves the issue, and email is sent to the receiver. All configuration happens in the Azure AI Foundry portal: selecting knowledge, managing agents, assigning tools, and adding the orchestration.
Attendee Takeaways

Multi-agent recipe – Guidance for getting started with specialist agents without even writing code.
Tool-selection – How to utilize Azure AI Search, Code Interpreter, Logic Apps, and other actions in agents
Business impact – A pattern you can adapt to various scenarios that needs a coordinated AI crew instead of a lone chatbot.

You walk away with knowledge of multi-agent solutions and how Azure AI Foundry Agent Service can be utilized.

Teams of Agents using Azure AI Foundry Agent Service

This session explores how Artificial Intelligence is reshaping the way humans work, communicate, and innovate. We will discuss the key challenges and opportunities of integrating AI into business, education, and public services—focusing on how young professionals can adapt and lead responsibly in an AI-driven world. The session will also include interactive discussion and examples of how AI tools are used in real-life scenarios

The Future of Human-AI Collaboration in the Digital Era

As the industry shifts from traditional prompt engineering to fully-fledged agentic AI systems, context management has become the new runtime. In this transformative landscape, the Model Context Protocol (MCP) emerges as a game-changer—empowering developers and architects to build intelligent, goal-driven agents that can reason, adapt, and act in dynamic environments.

In this power-packed session, delivered by a 4x Microsoft MVP, Global AI Influencer, and Chief Architect with deep expertise in Cloud, Data, and Generative AI, I’ll unpack how MCP is redefining context orchestration in modern AI systems.

Key takeaways from this session include:
• Architecture and Design Principles of MCP – how MCP structures and manages evolving context across AI workflows
• The role of MCP in AI Agent Development – enabling memory, tool usage, and dynamic state handling
• API vs MCP – choosing the right data strategy in agentic frameworks
• Seamless integration with Microsoft’s AI Stack – including Semantic Kernel, Azure AI Studio, and Copilot ecosystem
• Live demo – Build a secure and context-aware AI Assistant using Semantic Kernel and MCP
• Enterprise-grade focus on Security, Governance, and Compliance within MCP
• A forward-looking view of the future of MCP in scaling trustworthy and autonomous Gen AI applications

Whether you’re building multi-turn agents, designing enterprise copilots, or driving AI transformation at scale, this session will equip you with a blueprint to operationalise Model Context Protocol as the backbone of your agentic AI solutions.

The MCP Blueprint: Architecting Agentic AI

In this session, we’ll explore why traditional SharePoint security models fall short in today’s AI-driven landscape. How Sensitivity labels offer greater flexibility, control, and seamless integration with Microsoft Copilot, ensuring that content is protected at every touchpoint. We will finish by indicating how Sensitivity Labels strengthen compliance by enforcing data governance policies and auditability. This presentation is ideal for SharePoint professionals looking to modernise their SharePoint security framework and ensure compliance in an AI-powered environment.

Unlock Copilot with Sensitivity Labels security and AI Compliance .

In this session, Mohamed Touati will explore how Natural Language Processing (NLP) and Optical Character Recognition (OCR) are revolutionizing the healthcare industry. From automating clinical documentation to extracting insights from medical records, insurance forms, and handwritten reports, discover how AI can streamline workflows, reduce errors, and unlock the true potential of healthcare data. Real-world examples and practical frameworks will guide participants on building smart, scalable solutions for patient care, claims management, and beyond.

Share

Details

Address

Date / Time

Registration

Call for Speakers

Call for Sponsors

Social / Web

Hashtag

Website

Map

Community Days | AI DAYS TUNISIA