Discover what makes a successful AI agent in production (and how many of them fail on the way!) Learn about the key agentic design principles to set up your agents for scaling, including robust infrastructure and tooling, modular design architecture, and continuous evaluation and feedback loops.

AI Agents in the Wild

To agent or not to agent

Agentic applications

Design Principles for Scalable Agents

Strategies for scalable design

Enforcing the pillars of scalable design

How AI Agents Scale (and Fail)

Don't fail when you scale!

Agentic accounting

Guardrails or it fails

Designing Scalable Agents

Learn about key strategies to ensure that your agent is being developed with scalability in mind. Gain insights into how the Model Context Protocol (MCP) and the Agent-to-Agent protocol (A2A) enable scalability through standardization.

Multi-Agent Design Patterns

Making the leap from single to multi-agents

Multi-agent design patterns

The Model Context Protocol (MCP)

Getting to know MCP

MCP is the MVP

Clients, servers, and everything in between

The Agent-to-Agent (A2A) Protocol

Agent cards - Gotta Catch 'Em All

MCP and A2A: friend or foe?

Developing Agents for Scalability

Time for production, but not so fast! Build a robust testing framework to give you confidence that the AI agent will continue to perform in production. Choose the best deployment strategy for your use case, and learn how to integrate real-time data sources with your agentic system.

From Proof-of-Concept to Production

FoodGPT

Robustly testing an agent

Options

Tasks to complete

Not required

What's left to check?

Fast and Efficient Data Ingestion

The value of real-time data

Failing Gracefully: Mitigating Risks

Retry with exponential backoff

Laying down the law!

Congratulations!

Deploying Agents into Production at Scale

Cheat Sheet

Course Glossary

Why do so many agentic systems fail to traverse the gap between development and production? In this course, you'll find out! Learn how to design and develop AI agents with scalability in mind, following the three pillars of agentic scalability: robust infrastructure and tooling, modular design architecture, and continuous evaluation and feedback loops. You'll learn about how agentic frameworks like the Model Context Protocol (MCP) and the Agent-to-Agent protocol (A2A) enable cleaner and faster integrations between data sources and agents. Finally, learn about the tests and deployment strategies to set your agent up to deliver value.

The videos contain live transcripts you can reveal by clicking "Show transcript" at the bottom left of the videos.

The course glossary can be found on the right in the resources section.

To obtain CPE credits you need to complete the course and reach a score of 70% on the qualified assessment. You can navigate to the assessment by clicking on the CPE credits callout on the right.

<h2>Design and Develop Agents for Scaling</h2>
Learn how to design and develop AI agents with scalability in mind, following the three pillars of agentic scalability: modularity, robustness, and adaptability. Discover what makes a successful agent in production, and why so many struggle to get there.<br><br><h2>Discover the Power of MCP and A2A</h2>
The Model Context Protocol (MCP) developed by Anthropic has revolutionized agent interoperability, creating a unified approach for connecting agents to data sources. The Agent-to-Agent protocol (A2A) developed by Google compliments MCP. Find out how these two frameworks can be combined to ensure your agent's integrations are scalable.<br><br><h2>Implement Agent Testing and Deployment Best Practices</h2>Before pressing the big red button and launching your agent into production, you've got to mitigate the risks that come with scaling. Learn how to create a robust testing framework to capture issues with components, integrations, performance, and security. Decide which deployment type is right for your agent by looking at the needs of the use case.

Introduction to AI Agents

Discover what it takes to scale AI agents, with a little help from frameworks like MCP and A2A.

Building Scalable Agentic Systems

Fundamentos de agentes de IA

Ensure responsible guardrails and observability are in place

Remove cost and usage restrictions from tools

What's left to check?

Building Scalable Agentic Systems

exercicio interativo prático