Are Microservices Still Relevant in the Age of AI?

Four microservices experts — Mike Amundsen, Matt McLarty, Christian Posta, and Chris Richardson — argue that AI is not making microservices obsolete but is actually increasing their relevance. Key reasons include: microservices' foundational principles (clear boundaries, API-first design) align with AI system requirements like scalability and security; GenAI coding agents benefit from microservices constraints that limit scope of change and reduce LLM token costs; and microservices enable deterministic infrastructure that wraps probabilistic LLM behavior, reducing business risk. Tool-calling protocols like MCP reinforce the need for well-defined service interfaces and contracts in agentic architectures.