Marco Antonio Silva
Microsoft - Sr Global Black Belt - Solution Engineer - EMEA
Microsoft - Sr Global Black Belt - Solution Engineer - EMEA
Lisbon, Portugal
Actions
I’m a Senior Solution Engineer in the Global Black Belt team in EMEA, focussed on the Application Modernization space. This is a highly specialized team of people that help customers adopt cutting edge technologies, practices and Frameworks, while at the same time bringing important structured feedback directly from the field to Engineering teams on what needs to change in products to match customer's real need. Previously I was the National Innovation Officer for Microsoft Portugal, responsible for helping companies grow and accelerate the innovation impact that we can have in Portugal, having worked in Tech innovation for many years. I was also a Senior Cloud Solution Architect with many years of experience in the Telco Industry. As a Cloud Solution Architect my work was to challenge my customers with new ideas and new ways for them to use cloud technologies, and I love it.
I got my Engineering Degree in Computer Engineering from FEUP, specializing in Artificial Intelligence and Computer Graphics, which are all areas I love. I’m part of the AI Ambassador team at Microsoft so I’m tasked with the mission of empowering people to leverage AI more and achieve more with it. I’m a professional speaker for many years, and I teach some AI/ML classes at ISEG in Lisbon. I also host and founder of the “AI Portugal Podcast”, the "The Innovation Office Webshow" and the “DICE Podcast”, and the blog "Tech Alchemy".
Sou Senior Solution Engineer na equipa Global Black Belt na região EMEA, focado na área de Modernização Aplicacional. Esta é uma equipa altamente especializada que ajuda os clientes a adotar tecnologias, práticas e frameworks de ponta, enquanto traz feedback estruturado e importante diretamente do terreno para as equipas de Engenharia, indicando o que precisa de mudar nos produtos para corresponder às necessidades reais dos clientes.
Anteriormente, fui Responsável Nacional pela Inovação na Microsoft Portugal, com a missão de ajudar as empresas a crescer e acelerar o impacto da inovação que podemos gerar em Portugal, tendo trabalhado em inovação tecnológica durante muitos anos. Fui também Arquiteto Sénior de Soluções Cloud, com vasta experiência na indústria das telecomunicações. Como Arquiteto de Soluções Cloud, o meu trabalho era desafiar os clientes com novas ideias e novas formas de utilizarem tecnologias cloud — e adoro fazê-lo.
Licenciei-me em Engenharia Informática pela FEUP, com especialização em Inteligência Artificial e Computação Gráfica, áreas pelas quais tenho grande paixão. Faço parte da equipa de AI Ambassadors na Microsoft, com a missão de capacitar as pessoas para tirarem maior proveito da IA e alcançarem mais com ela. Sou orador profissional há muitos anos e leciono algumas disciplinas de IA/ML no ISEG, em Lisboa.
Sou também host e fundador do podcast “AI Portugal Podcast”, do programa “The Innovation Office Webshow”, do “DICE Podcast”, e do blog “Tech Alchemy”.
Area of Expertise
Topics
Spec-Driven Development with GitHub Copilot & Spec2Cloud: From Idea to Cloud — Green & Brownfield
Most teams either start from a blank file and guess the architecture as they go (Greenfield), or they stare at a legacy codebase and hope GitHub Copilot figures out what it should become (Brownfield). Both approaches leave enormous value on the table. Spec-Driven Development changes the game: you define the contract first, then let Copilot and Spec2Cloud generate, validate, and deploy the implementation.
In this full-day hands-on workshop, you will learn and practice the full Spec-Driven Development workflow using GitHub Copilot and Spec2Cloud — covering both Greenfield and Brownfield scenarios. In the morning we build a new cloud-native application from scratch: starting with an OpenAPI spec authored with Copilot, generating the implementation skeleton with Spec2Cloud, and iterating with Copilot agents until we have a tested, deployable service. In the afternoon we tackle Brownfield: we take a real legacy application, use Copilot to reverse-engineer its implicit contract into an explicit spec, then drive a structured modernization using Spec2Cloud to produce a cloud-ready target.
By the end of the day you will have hands-on experience with both flows, a reusable methodology for your team, and working code you can adapt immediately.
✅ Hands-on from minute one — real code, not toy examples
✅ Two complete flows: Greenfield build + Brownfield modernization
✅ Take-home repo with exercises, solutions, and a methodology reference card
The Age of Coding Agents: How AI is Rewriting the Way We Build Software
We are living through the most profound shift in software development since the invention of version control. AI is no longer your autocomplete — it is your coding agent. It plans, executes, reviews, and ships code autonomously. The question is no longer "should we use AI?" but "how do we build an engineering culture that harnesses agents at scale?"
In this keynote, Marco Antonio Silva — Microsoft Global Black Belt and field engineer who runs agentic development programs with Fortune 500 enterprises — takes you on a live journey through the three pillars of the Coding Agent revolution:
🤖 GHCP Modernization CLI — Watch a legacy codebase get analysed, planned, and refactored by an agent in minutes. Real code. Real enterprise context. The same workflow being used today to modernize hundreds of applications at scale.
🛠️ Squad — Meet the next evolution: multi-agent teams that collaborate the way engineers do. Squad brings structured agentic ceremonies (planning, review, merge gates) to your repository, transforming a single AI into an always-on engineering crew.
🚀 Spec2Cloud — See how a simple specification becomes a cloud-native application. Spec2Cloud closes the loop between intent and implementation, bridging the gap from design to deployed service with agents doing the heavy lifting.
This is not a future vision. Every demo in this keynote runs live. Every pattern shown is in production at enterprise customers today. You will leave with a concrete mental model for how agents change the software lifecycle — and what you need to do next week to get ahead of this shift.
✅ Understand the three layers of the Coding Agent revolution
✅ See real agents modernize, build, and deploy code live on stage
✅ Walk away with an actionable adoption roadmap for your team
Level Up your Repository with Agentic AI: Spec-Driven Development with GitHub Copilot
Most developers use GitHub Copilot as a fast typist. The ones who have rethought their workflow use it as an autonomous collaborator that designs, implements, reviews, and refines — with the developer in the role of architect, not coder. That shift starts with Spec-Driven Development.
In this session, you will see how to "level up" your repository by introducing a spec-first workflow powered by GitHub Copilot and Spec2Cloud. Instead of generating code from vague prompts, you define an explicit contract — an OpenAPI specification, an architecture doc, or a structured requirements file — and let Copilot agents use that spec as ground truth for every code change, test, and refactoring decision.
We go further with Squads: a new multi-agent development approach where specialized AI agents (designer, coder, reviewer, tester) collaborate in your repository on well-defined tasks, coordinated by a spec that keeps them aligned. You will watch a feature go from specification to reviewed pull request — with multiple agents contributing — live on stage.
🗂️ Understand why spec-first is the foundation of effective agentic development
✍️ Use GitHub Copilot to author, iterate, and validate OpenAPI and architecture specs
🤖 Set up Squads: coordinated multi-agent workflows in your repository
🔄 Build a workflow where agents handle implementation while you own the architecture
🚀 Apply Spec2Cloud to generate cloud-ready scaffolding directly from your specs
If you write code for a living and want to stop prompting and start architecting, this session will change how you work — starting tomorrow.
Autonomy Unleashed: Multi-Agent Systems from MAF Orchestration to Squads
Single agents are powerful. But the real breakthrough comes when agents collaborate — each one specialised, each one autonomous, all working together toward goals no individual agent could tackle alone. The challenge is that multi-agent architectures have been evolving fast, and developers are confronted with a growing landscape of patterns and tools.
In this session, we navigate that landscape together — from structured orchestration with Microsoft Agent Framework (MAF) to the emerging Squads paradigm: a new approach to agentic development where specialised squads of agents tackle complex tasks through composable, squad-based workflows.
We will cover:
🤖 The case for multi-agent systems: when and why to go beyond a single agent
🏗️ Core MAF concepts: Agent roles, communication channels, and orchestration patterns
🔄 Coordination patterns: sequential, parallel, and event-driven agent pipelines
🧩 Squads: the new paradigm for composable, squad-based agentic development
🔒 Governance and observability: keeping multi-agent systems trustworthy and debuggable
🌍 Real-world patterns from enterprise deployments in banking, telco, and insurance
We'll walk through a live demo of a multi-agent pipeline — built first with MAF, then evolved into a squad-based workflow — handling a complex, multi-step business process entirely through agent collaboration.
You'll leave with a clear mental model for designing multi-agent systems, an understanding of where MAF and Squads each shine, and code patterns you can apply immediately.
GitHub Copilot Deep Dive: From Zero to Hero with AI-Powered Development
GitHub Copilot has moved far beyond autocomplete. In 2026, it is an autonomous coding agent, a code reviewer, a test generator, a documentation writer — and for developers who know how to use it properly, a genuine force multiplier. But most developers are only using 20% of what it can do. This session changes that.
In this high-energy, demo-driven session, we go from zero (what even is Copilot?) to hero (autonomous agent mode, custom instructions, multi-file edits, and workflow integration) in 60 minutes. You will watch a real .NET application get built, tested, and documented faster than you thought possible — with Copilot as a collaborator, not just a suggester.
We cut through the hype and focus on what actually works in a professional development workflow: how to write prompts that produce great results, how to use Copilot Workspace for larger features, how to set up custom instructions that make Copilot fit your team's standards, and how to integrate Copilot into your CI/CD pipeline for automated code review and test generation.
🚀 What you will learn:
✅ The full Copilot toolkit in 2026: from completions to autonomous agents
✅ Prompt engineering techniques that actually produce quality code
✅ Custom instructions: making Copilot follow your team's standards automatically
✅ Copilot for tests, docs, and code review — beyond just writing code
✅ How to build a Copilot-augmented workflow your entire team will adopt
Agentic Modernization: Transforming Legacy Applications with GitHub Copilot CLI
Every organization has legacy applications — and most modernization projects fail not because of technology, but because teams cannot keep up with the analysis, planning, and execution at scale. What if your AI pair programmer could lead the modernization instead of just assisting it?
In this session, you will see how GitHub Copilot CLI turns the modernization workflow on its head. Rather than manually analyzing codebases, writing migration specs, and hand-crafting refactoring plans, you use Copilot's agentic capabilities to orchestrate the entire process: scanning legacy code for patterns and risks, generating structured migration proposals, executing refactoring steps autonomously, and validating the results — all from the command line.
Drawing on real enterprise modernization programs (including a major European bank with 150+ applications in flight), this session shows what agentic modernization looks like in practice: where it accelerates the work dramatically, where human judgment is still essential, and how to build a repeatable workflow your team can apply at scale. You will leave with a concrete methodology, live demo code, and the patterns needed to start applying agentic modernization in your own organization.
🤖 What you will learn:
✅ How GitHub Copilot CLI's agent mode works for autonomous code analysis and transformation
✅ A repeatable agentic modernization workflow: scan → plan → execute → validate
✅ Real-world patterns from modernizing 150+ applications at a European enterprise
✅ Where to trust the agent and where to keep a human in the loop
✅ How to build custom Copilot CLI extensions to encode your team's modernization rules
Autonomy Unleashed: Building Production-Ready Multi-Agent Systems with Agent Framework & Aspire
AI agents are no longer a research curiosity — they're shipping to production today. But most teams hit the same wall: a promising prototype that collapses under real-world complexity. Orchestrating multiple agents, managing state, handling failures gracefully, and integrating with enterprise systems turns out to be far harder than the demo suggested.
In this session, we go beyond the "hello world" agent and build something real. Using the Microsoft Agent Framework and .NET Aspire, we design a distributed multi-agent system that tackles a genuine enterprise modernization scenario — live, on stage, with code you can take home.
You will see how to compose specialized agents into reliable pipelines, how to coordinate agent collaboration and state across a distributed system, and how Aspire's orchestration and observability layer makes deploying and monitoring multi-agent workflows dramatically simpler. We cover the patterns that actually hold up in production: structured handoffs, tool grounding, observability with OpenTelemetry, and graceful degradation.
🤖 Understand the architecture behind production multi-agent systems with Microsoft Agent Framework
🔧 Design and wire specialized agents with well-defined roles, tools, and handoff contracts
🔗 Integrate agents with enterprise data sources and Azure services
⚙️ Use .NET Aspire to orchestrate, observe, and deploy distributed agent workflows
🛡️ Design for failure — retries, fallbacks, and human-in-the-loop checkpoints
Whether you are an AI engineer ready to ship your first agent or a developer curious about what autonomous AI systems really look like under the hood, this session delivers production-grade clarity — no hype, all code.
Agentic Modernization Lab: Modernize Legacy Repos with GitHub Copilot Modernization CLI & Spec2Cloud
Millions of lines of legacy code sit dormant across enterprises — costly to maintain, impossible to evolve, and a barrier to cloud adoption. What if AI agents could do the heavy lifting of modernizing them?
In this hands-on workshop, we tackle the full Application Modernization pipeline using two powerful tools: GitHub Copilot Modernization CLI for AI-driven automated code analysis and transformation, and Spec2Cloud for spec-enabling legacy repositories through Spec-Driven Development. Together, they form an agentic modernization pipeline that turns legacy codebases into cloud-native, spec-enabled, maintainable systems — with minimal manual effort.
Every attendee works on a real legacy repository and takes it through the full modernization journey:
🔍 Analyse: Run GitHub Copilot Modernization CLI to assess the codebase, identify modernization candidates, and generate a transformation plan
📋 Spec-Enable: Apply Spec-Driven Development principles to generate API contracts and specifications from legacy code using Spec2Cloud
🤖 Agentic Transformation: Use AI agents to apply spec-driven refactoring patterns, modernize dependencies, and update configuration for cloud targets
☁️ Cloud-Ready: Validate the modernized codebase against cloud-native patterns and prepare it for container or serverless deployment
🔄 CI/CD Integration: Wire the modernization pipeline into GitHub Actions for continuous modernization at scale
Attendees will leave with a modernized repository, a reusable agentic pipeline, and the skills to apply this workflow to their own legacy systems.
Prerequisites: Laptop with Git, GitHub CLI, and .NET or Java SDK installed. A GitHub account with Copilot access. Setup instructions provided in advance.
MAFSpire: Using Microsoft Agent Framework and Aspire together
In this day-long workshop, you’ll be learn how to build custom agents using Microsoft Agent Framework (MAF) and how to distribute that agent topology across multiple nodes in a distributed environment using Aspire. You’ll learn how to build custom instructions that tailor your agents’ behaviour, as well as building complex workflows and agent-to-agent communication. You’ll walk away from this workshop with a deep knowledge of how to build and deploy agent-based systems using Aspire, Azure AI Foundry, and Microsoft Agent Framework.
AI Meets Platform Engineering: Automating Modernization with Radius
Modernizing thousands of enterprise applications is complex and error-prone. In this session, Millennium bcp and Microsoft show how AI copilots and Radius provide an app-centric foundation for platform engineering at scale.
We’ll demonstrate how AI-assisted workflows analyze legacy workloads, generate Radius app graphs, and automate Kubernetes deployments, while enforcing policies and lifecycle management. Attendees will see how AI augments platform teams, reduces manual toil, and enables consistent modernization across hybrid and multi-cloud environments.
The talk covers architecture, automation patterns, and lessons learned from running AI-driven modernization pipelines in production—helping participants understand how Radius and AI together make modernization practical, explainable, and open-source friendly.
🚀 Launching at Scale: Platform Engineering Made Easy with Radius
Cloud Native success doesn’t just happen — it’s engineered. In this session, we’ll explore how Radius by Microsoft empowers teams to scale best practices, standardize environment setup, and embed platform engineering best practices from the start.
🔍 What You’ll Learn:
📦 What is Radius?
A guided tour of Microsoft’s open-source Cloud Native application definition framework
🧱 Building Blocks for Platform Engineering
How to abstract complexity and enable developer self-service without compromising control
🚦 Standardizing Environment Setup
From dev to production — repeatable, reliable, and secure
🌍 Scaling Across Teams and Clouds
Best practices to scale governance, observability, and Platform Engineering across your organization
Whether you're starting your cloud native journey or scaling to support dozens of teams, this session will equip you with the insights and tools to succeed in your journey— with Radius at the core. 🌟
GitHub Copilot Deep Dive: From 🛠️ Zero to 🦸 Hero, with your 🤖 AI Pair Programming buddy
Ready to turbocharge your development workflow? In this hands-on, fast-paced session, we’ll explore how GitHub Copilot is much more than just code autocomplete — it’s your full-on AI-powered development partner. Whether you're just getting started or looking to master every feature, this session will take you from 🛠️ Zero to 🦸 Hero.
👨💻 What You’ll Learn:
✍️ IDE Magic (VS Code & beyond):
Discover how Copilot helps you write code faster, smarter, and with fewer bugs — right from your editor.
🔍 Smart Suggestions in Context:
Learn how Copilot understands your codebase and offers context-aware completions that actually make sense.
📦 GitHub Portal Superpowers:
From AI-powered pull request reviews to autofix suggestions, see how Copilot lives inside your GitHub workflows.
🧠 Prompting Tips & AI Best Practices:
Write better prompts and get better results — unlock the true potential of Copilot’s capabilities.
⚙️ Copilot Labs & What’s Coming Next:
A sneak peek into advanced features and experimental tools shaping the future of AI development.
💥 Why You Should Join:
Whether you’re coding solo or as part of a team, GitHub Copilot can help you write better code, faster — and free up your time for more creative and impactful work. Walk away with practical tips, real examples, and a fresh perspective on what it means to code with AI.
🎯 Audience:
Developers of all levels who want to get more done, learn faster, and level up their daily workflow with AI.
MEO: an AI-Powered, Business-Driven Techco
• How to setup a business-driven strategy on a Telco Provider
• Target operations as a low-hanging fruit for improving efficiency and customer experience
• Use your most demanding customer painpoint as the energy to leverage AI strength
• Collect the results as improvements on your customer experience
Panel: Cloud Careers in the Age of AI: Navigating the Future of Work
As artificial intelligence reshapes the digital landscape, cloud professionals stand at the forefront of one of the most transformative eras in technology. This panel explores how AI is redefining roles, skills, and career paths within the cloud ecosystem. Industry experts will discuss the evolving demands of cloud jobs, the fusion of AI and cloud technologies, and how professionals can future-proof their careers through continuous learning and strategic upskilling.
🚀 Autonomy Unleashed: Agent Framework and the New Age of AI Agents 🤖✨
Step into the future of AI! 🔮 In this dynamic session, we’ll explore how Microsoft Agent Framework empowers a new generation of agentic AI 🤝 — systems that don’t just respond, but reason, plan, and act 🧠📅🛠️. From orchestrating complex tasks to adapting on the fly, these smart agents are changing the game 🎮 across industries.
Get inspired 💡, see real-world magic 🪄, and discover how you can start building AI that truly gets things done ✅🔥.
🚀 Unleashing AI Agents: A Live Demo with Semantic Kernel 🤖
Curious about how AI agents can take action in real-world applications? In this fast-paced, demo-driven session, we’ll showcase Microsoft Semantic Kernel in action—no slides, just code and live demos!
You’ll see first-hand how Semantic Kernel enables AI models to reason, plan, and execute functions autonomously by connecting them to APIs and existing business logic. We’ll walk through real-time examples of AI agents dynamically responding to user inputs, calling external services, and executing multi-step tasks—all in just 15 minutes.
What you'll see live:
🚀 An AI agent making real-time decisions and executing tasks
🔗 Connecting AI models to APIs and existing code with Semantic Kernel
🤖 Automating workflows using Agentic AI principles
💡 Practical takeaways to start building AI-powered applications today
If you want to see AI agents in action, this session is for you!
Microsoft Project Orleans in Action: Building Scalable and Resilient Distributed Apps
Microsoft Project Orleans simplifies the complexities of distributed application development, allowing developers to build robust, cloud-native, and scalable solutions with ease. In this hands-on session, we will dive deep into Orleans' core concepts, architecture, and benefits through a practical lens.
Through live demos and real-world use cases, we will explore how Orleans enables effortless scalability, fault tolerance, and stateful service orchestration across distributed environments. Whether you're new to distributed systems or looking to leverage Orleans for your next cloud-native project, this session will provide the insights and practical knowledge needed to get started.
Key takeaways:
✅ Understanding Orleans' unique actor-based model and virtualized grains
✅ Deploying Orleans applications across Kubernetes, VMs, and cloud platforms
✅ Scaling elastically and handling failure scenarios seamlessly
✅ Hands-on demos showcasing real-world implementations
Join us to see Project Orleans in action and discover how it can transform the way you build and scale distributed applications!
Today’s Innovation is Cloud Native
There is no question that the world has changed to the cloud. This became more and more noticeable with the growth of AI tools, and more and more Cloud Native services and products become part of every modern application design. With the rise of all copilots on our day by day, from Developers, to Operations, to IT Pros. The landscape has permanently changed, and the potential is immense, and exciting to imagine how we can do so much more with all of these new tools and services.
AI Panel Discussion: Navigating the AI Frontier
Join us in this panel discussion, where industry experts explore the latest trends, challenges, and breakthroughs in artificial intelligence powered by Azure.
From machine learning to cognitive services, our panelists will share their insights on leveraging Azure's cutting-edge technologies to build intelligent solutions that drive innovation and transform industries. Discover real-world use cases, best practices, and the future landscape of AI on the Azure platform.
Whether you're a developer, data scientist, or business leader, this session offers a unique opportunity to gain valuable perspectives on the evolving AI landscape within the Azure ecosystem.
Deep Dive into AKS
The world runs on the Cloud, it's a know fact by now. And the cloud runs on K8! So come and let's deep dive into AKS Architectures, patterns, recipes, best practices and a lot of Kubectl and helm and all of these cool toys. This is not your run of the mil AKS session, it's an advanced session to get you over the hurdle on how to use AKS in enterprise grade scenarios.
Marco Antonio Silva
Microsoft - Sr Global Black Belt - Solution Engineer - EMEA
Lisbon, Portugal
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top