API Superstream: APIs and Agents
Published by O'Reilly Media, Inc.
Designing APIs for the demands of AI
Traditional APIs are designed for human interaction, but AI agents mark a fundamental shift in development patterns. Forward-thinking API design must anticipate agents that can independently discover endpoints, authenticate themselves, execute precise operations, and gracefully disconnect. This architectural evolution requires rethinking everything from rate limiting and authentication to documentation and error handling. So how do you prepare?
Join host Mike Amundsen and an esteemed lineup of API experts to learn how to create APIs that are both human-accessible and agent-optimized, and explore the opportunities and challenges that lay ahead.
What you’ll learn and how you can apply it
- Explore the current trends at the intersection of AI and APIs and be ready to apply them to your work
- Learn how to build APIs that are compatible with AI
- Discover the ways AI can impact user engagement, trust, and transparency and what you can do to mitigate these issues
Recommended follow-up
- Take AI-Driven API Design (live course with Mike Amundsen)
- Read AI Engineering (book)
- Watch Coding with AI: The End of Software Development as We Know It (event video)
- Read Hands-On APIs for AI and Data Science (book)
Schedule
The time frames are only estimates and may vary according to how the class is progressing.
Introduction – Mike Amundsen (5 minutes)
Mike Amundsen welcomes you to the API Superstream.
APIs for AI: Have We Failed? – Zdenek Nemec (35 minutes)
APIs have become a significant bottleneck for autonomous agents. With the growth of AI agents, the need for suitable APIs is greater than ever. As API providers, we often don’t react quickly enough to this demand, limiting the application of AI in real-world scenarios. Zdenek Nemec, founder and CTO of Superface.ai, explores the issues with current APIs from the standpoint of LLM-powered AI agents, and discusses best practices for AI-appropriate APIs as well as remedies and alternatives when APIs are unsuitable for AI use.
Building and Securing Remote MCP Servers Directly from Your API Gateway – Martyn Davies (35 minutes)
Martyn Davies, developer advocate at Zuplo, shows you how to use your existing APIs to generate and host remote Model Context Protocol servers to allow AI access. He’ll explore the role API gateways play in hosting and securing your MCP server and how they can help protect you from common AI/MCP vulnerabilities.
Break (5 minutes)
Built for Agents or Built to Break? Ready Your APIs – Markus Mueller (Sponsored by Boomi) (30 minutes)
Agentic transformation runs on APIs—the bridge that AI agents require to access data and services and to collaborate. To support trusted agentic AI, APIs must be governed, secure, and built for mission-critical performance. Markus Mueller, Boomi field CTO, explores how Boomi API Management enables enterprise-grade agentic ecosystems through federated API governance, standardized API design workflows, built-in API protection, secure scalability, and Model Context Protocol (MCP) support. Learn how to ensure that your APIs are fast, scalable, and secure.
This session will be followed by a 30-minute Q&A in a breakout room. Stop by if you have more questions for Markus.
Bridging APIs and AI: LLMs, GenAI, and the Future of XAI Systems – Manai Mohamed Mortadha (35 minutes)
APIs are the backbone of modern AI, enabling seamless integration of powerful technologies like large language models and generative AI into real-world applications. As an AI engineer at Netflix and an XAI researcher, Manai Mohamed Mortadha has witnessed firsthand how APIs power scalable AI systems, from personalized content recommendations to innovative explainability features. Hel explores the synergy between APIs, LLMs, and XAI, highlighting their transformative impact on user engagement, trust, and transparency.
Break (5 minutes)
Empowering Your Agents: APIs as Tools with Azure API Management and MCP – Julia Kasper (35 minutes)
You can unlock the full potential of your AI agents by seamlessly integrating your APIs as powerful tools. Julia Kasper, a program manager at Microsoft, demonstrates how to leverage Azure API Management to expose your existing APIs, transforming them into accessible and reliable functionalities for your agents. You’ll dive into the Model Context Protocol (MCP), and learn how it facilitates effective communication and utilization of these API tools, enabling your agents to perform complex tasks and make informed decisions with greater autonomy and precision.
Using APIs with AI – Ryan Day (35 minutes)
In technology circles, AI and APIs are sometimes treated as separate specialties. But they are closely related and getting closer all the time. Advanced data scientist Ryan Day explores the ways that AI and APIs overlap, some of the skills you may need to develop, and how to build APIs that are compatible with AI.
Closing Remarks – Mike Amundsen (5 minutes)
Mike Amundsen closes out today’s event.
Your Hosts and Selected Speakers
Mike Amundsen
Internationally known author and speaker Mike Amundsen consults with organizations around the world on network architecture, web development, and the intersection of technology and society. He helps companies capitalize on the opportunities provided by APIs, microservices, and digital transformation.
Zdenek Nemec
Zdeněk “Z” Němec is the founder and CTO of Superface.ai, where he is pushing the boundaries of agentic AI. He also founded Good API, helping enterprises such as DHL, Adidas, Inditex, and KLM-AF build high-quality APIs faster. And, as the former director of DSL development at Apiary, he authored the API Blueprint language, which has been used by over 200,000 developers worldwide to describe more than 300,000 APIs.
Martyn Davies
Martyn Davies is a developer advocate at Zuplo with years of experience leading developer advocacy, education, community management, and product management at companies large and small. He has built, led, and enjoyed being part of forward-thinking advocacy teams since 2012.
Markus Mueller
Markus Mueller is field CTO at Boomi, where he leads strategic initiatives in API management and AI agent integration. With deep expertise in federated API governance, scalable architectures, and emerging AI technologies, he helps enterprises build secure, agent-ready API ecosystems. Previously, Markus was cofounder and CTO of APIIDA. He’s also a frequent speaker on API strategy and digital transformation at international conferences.
Manai Mohamed Mortadha
Manai Mohamed Mortadha is a senior AI engineer at Netflix, where he works on building explainable AI (XAI) systems that drive content recommendations, optimize platform decisions, and uphold ethical standards in model deployment. Also, as founder and CEO of Man.AI, he leads the development of responsible AI solutions for enterprises worldwide with a strong focus on large language models (LLMs), retrieval-augmented generation (RAG), AI governance, and scalable ML systems in healthcare, finance, and education.
Julia Kasper
Julia Kasper is a program manager at Microsoft. Her work on Azure API Management and API Center directly supports the Model Context Protocol by helping organizations catalog, manage, and govern their APIs/MCP servers for easier discovery by intelligent agents. She’s keen to explore how new API trends can enhance developer productivity and enjoys connecting with those who share her passion or want to discuss Azure and its relevance to MCP.
Ryan Day
Ryan Day is an advanced data scientist in the finance industry. Previously, he led the digital services division for a federal agency, advancing cloud computing, data science, and API development initiatives. Ryan is an experienced Python and open source developer and a certified AWS Solutions Architect.
