Archive — US — techcrunch
Dapr’s microservices runtime now supports AI agents
Microsoft's Dapr now supports AI agents, enhancing its microservices runtime to facilitate AI agent development. Dapr's core feature of virtual actors allows scalable and resource-efficient Agent management, beneficial for orchestrating AI systems. Dapr Agents, built on the open-source Floki project, aim to streamline the development process with integrated statefulness and orchestration, currently supporting major model providers.
Key points
- Dapr is a microservices runtime introduced by Microsoft in 2019, initially not focused on AI.
- Dapr's core feature, virtual actors, allows independent message processing, supporting AI agent frameworks.
- Dapr Agents offer a streamlined method to develop AI agents with built-in orchestration and state management.
- The project originated from Floki, enhancing Dapr's capabilities for AI agents and ensuring continuity.
- Currently supports popular LLMs like AWS Bedrock and OpenAI, with local LLM support coming soon.
- Dapr Agents are being rolled out for Python, with .NET, Java, JavaScript, and Go support following.
Why it matters
- The rise of AI and machine learning technologies has made orchestration and state management crucial in software development.
- Historically, distributed systems have evolved with innovations like microservices; AI agents represent a natural progression in this field.
- The integration of AI into microservices frameworks aligns with 2025 IT trends emphasizing automation, efficiency, and the use of AI-driven solutions across industries.
Dapr's support for AI agents marks a significant evolution in microservices, reflecting the growing interplay between distributed systems and AI capabilities while providing developers with essential tools to build efficient, scalable AI applications.