Why it matters
In the workshop you'll build a team of AI agents from scratch, deploy them to the cloud, coach them through live matches, and iterate on their performance using agentic coding tools. That's not a demo. That's a real multi-agent system running on production-grade AWS infrastructure.
Take a step back and look at what you'll actually be doing:
Every one of those skills maps directly to real-world agentic AI applications. The football pitch is just the sandbox. The patterns are the same ones powering production systems across industries today.
Amazon Bedrock and AgentCore aren't just for football. Foundation models, managed agent runtimes, tool integration, memory, observability: the services you'll use in the workshop are the same ones teams rely on to build production agentic systems that solve real business problems.
And the same agentic coding workflow you'll use to iterate on your football agents? That works for any agent project. Open Kiro, describe what you want to build, and let it help you scaffold, code, deploy, and debug.
These aren't hypothetical use cases. Organizations across industries are already running agentic AI in production on Amazon Bedrock and AgentCore, using the same stack you will.
The global pharmaceutical company built a multi-agent "Development Assistant" on Amazon Bedrock that enables clinical, regulatory, safety, and quality teams to query and analyze clinical trial data through a conversational interface. What used to take days of manual data gathering now happens in seconds, accelerating their goal of delivering 20 new medicines by 2030.
Read the AstraZeneca case study (opens in a new tab)Built Review Resolve™ on AgentCore, an AI copilot that analyzes clinical records, patient notes, and faxes to validate medical necessity for prior authorization requests. They expect to reduce review times by 30-40% while improving their existing 90% automation rate, all in a highly regulated healthcare environment where every minute matters for patient care access.
Read about Cohere Health on AgentCore (opens in a new tab)Transformed financial research for millions of investors using a multi-agent system on Amazon Bedrock AgentCore. Specialized agents handle structured data (stock prices, financials), unstructured data (SEC filings, news), and various APIs, all coordinated through a supervisor-subagent pattern.
Learn about Yahoo Finance's multi-agent system (opens in a new tab)Australia's largest fintech built Guardian, an agentic AI application on Amazon Bedrock that monitors homeowner loans around the clock, provides personalized refinance insights, and simplifies the mortgage process. They went from concept to production in just 12 weeks.
Read how Lendi built Guardian (opens in a new tab)The world's largest automotive services provider launched five agentic AI products using AgentCore and the Strands Agents SDK, including a fully automated virtual assistant for dealership customer conversations that operates autonomously after hours. Their agents streamline consumer car shopping, fleet service, vehicle auctions, and dealership operations.
Read the Cox Automotive case study (opens in a new tab)Part of the world's largest advertising company, Epsilon uses AgentCore to power their Intelligent Campaign Automation solution. AI agents automate campaign design, audience targeting, and real-time optimization across multiple channels, delivering faster execution and improved targeting precision at scale.
Learn about Epsilon's campaign automation (opens in a new tab)One of the world's largest utility companies deployed multi-agent systems on AgentCore to transform IT operations in ServiceNow. Orchestrated agents validate change requests, enrich incident tickets with contextual intelligence, and guide teams through change model selection via conversational AI. The result is fewer bottlenecks and faster ticket resolution across the organization.
Read how Iberdrola uses AgentCore (opens in a new tab)The global agriculture leader built Cropwise AI using Amazon Bedrock Agents, an AI assistant that supports sales representatives with precision farming recommendations. The system helps farmers optimize crop management decisions based on real-time data, marking a new era in agricultural technology.
Read about Syngenta's Cropwise AI (opens in a new tab)Uses AgentCore to develop an agentic manufacturing approach where AI agents work together using product specifications to automate manual processes. One agent reads product requirements and creates test procedures for quality control, while another trains vision systems for manufacturing robots. Fine-tuning an object detection model went from days of engineering time to under an hour.
Read about agentic manufacturing (opens in a new tab)The global telecom leader uses AgentCore across systems spanning millions of lines of code and thousands of interconnected subsystems. Their agents deliver double-digit productivity gains across a workforce in the tens of thousands, fusing data and information to deliver unprecedented capability in real-world R&D.
Learn about Ericsson's agent deployment (opens in a new tab)Built a group-wide Agentic AI Platform on AgentCore, achieving enterprise-level security, observability, and scalability with seamless cross-platform connectivity to AI resources. Sony describes AgentCore as the core of their agentic AI ecosystem, enabling them to govern and share AI capabilities across the organization.
Read about Sony's agentic AI platform (opens in a new tab)The AgentCore SDK has been downloaded over a million times. Early adopters also include Clearwater Analytics, Druva, Experian, Heroku, National Australia Bank, and Thomson Reuters, with support from partners like Accenture, Cisco, Deloitte, and Salesforce.
Two ways in. Bring the workshop to your team, or join the waitlist for the next virtual league season.
The loop is the same one you'll master on the pitch: build, deploy, observe, improve. The only difference is that the pitch becomes your production environment.
Come to play football. Leave with the skills to build production agentic AI.