Microsoft previews open-source Agent Framework for .NET and Python

🛠️ Microsoft Agent Framework: open source is where agents grow up. Microsoft just unveiled the Agent Framework (preview) — an open-source dev kit that unifies the best of Semantic Kernel and AutoGen into one powerhouse for AI agent creation. Why it matters for the open-source community: 🌍 Supports .NET & Python for maximum accessibility 🔗 Built on open standards (MCP, A2A, OpenAPI) 🔄 Orchestration patterns from simple sequential tasks → complex multi-agent collaboration 🧩 Community-first approach — Microsoft explicitly wants contributions This isn’t just another framework drop. It’s the recognition that agents are becoming the new application layer, and open source is the foundation that ensures durability, transparency, and trust. 💡 Question for the community: How would you like to see the Agent Framework evolve in the open-source ecosystem?

🌟 Unlock the power of multi-agent systems with the Microsoft Agent Framework! This open-source SDK allows developers to build, orchestrate, and deploy agents seamlessly in .NET and Python, integrated with Azure. Whether you're creating conversational agents or complex workflows, our graph-based orchestration and extensible abstractions ensure you can innovate rapidly and securely. Best of all, it's FREE! Explore how you can enhance your AI capabilities today. Learn more: https://lnkd.in/g2UuiWMN

👑 “The king is dead, long live the king.” I’ve been using Semantic Kernel for a while…and quite excited about what Microsoft just announced — the Semantic Kernel is transforming into the new Microsoft Agent Framework. In short — it’s not a goodbye, it’s a level-up. 🚀 Key promises: 🔹 Cleaner orchestration — sequential, concurrent, and dynamic agent handoffs are now built-in. 🔹 Stable, GA APIs — Agents are now Generally Available for .NET and Python. 🔹 Cooperation with AutoGen — cutting-edge experimentation meets Microsoft-grade stability. 🔹 Enterprise focus — fewer breaking changes, clearer telemetry, and more predictable updates. Having worked with SK for multi-agent workflows, tool chaining, and contextual memory, I can say this update promise to addresses real developer pain points. Less boilerplate, more power, and better composability (a bit to marketing, but hands on code will show it 😁) The way we design agentic systems in .NET is about to get simpler, cleaner, and much more scalable. 💭 My advice? If you’ve built custom agents or copilots already — take a weekend to explore the new framework. This is where enterprise-grade agent architecture is heading. What is your opinion about this level up? Let’s give it a try and see it in action: https://lnkd.in/dGBW9B3S #net #csharp #dotnet #aspnet #github #copilot #innovation #semantickernel #microsoft #AIagents #developers #softwareengineering #autogen #agentframework #AI #MachineLearning #artificialintelligence #innovation

Microsoft just released Agent Framework, an open-source toolkit that lets developers build AI helpers for .NET and Python projects. You can create a smart assistant that reviews code, schedules meetings, and handles customer queries without switching between different tools. **So what?** Small development teams get enterprise-level AI automation without massive budgets. Build your first AI agent → https://lnkd.in/g2-kCcjM

Continuing my portfolio series, where I tackle real-world business challenges with AI. Project 3: "If I was an AI Engineer at Microsoft... I'd build the Legacy Linter Ultimate." The Problem: Technical debt from legacy code is a major roadblock to innovation for enterprise companies. This old, undocumented code is risky to update and expensive to maintain. My Solution: I developed the "Legacy Linter Ultimate," an intelligent web application that de-risks and accelerates code modernization. It doesn't just refactor code; it uses AI to generate a complete "Code Modernization Package" that includes: * Automated, line-by-line documentation and professional docstrings. * Actionable suggestions for performance and security improvements. * Auto-generated, runnable unit tests for the new code. * An AI-generated flowchart to visualize the new code's architecture. * It also provides a detailed analysis of the idea behind your legacy code so you can view your legacy code through the eyes of the creator to understand what they had in mind when writing the code. This is the type of developer tool that would be a powerful addition to the Microsoft ecosystem, fitting perfectly with platforms like GitHub and Azure to help enterprise clients accelerate their digital transformation. It could also be used as an extension app in vscode if integrated properly. Tech Stack: GPT-5, Python, FastAPI, Next.js, TypeScript, Vercel I'm passionate about building AI tools that empower developers. I invite you to see the live demo: [Link to the live Vercel app: [https://lnkd.in/dEXBVjXn]] Disclaimer: This is a personal portfolio project, originally designed for an IBM Hackathon. It is for educational and skill-demonstration purposes only and is not affiliated with Microsoft, IBM, or any other platforms mentioned. #AI #SoftwareEngineering #DeveloperTools #Microsoft #GitHub #TechnicalDebt #Python #FastAPI #PortfolioProject #GenAI #IBM #Microsoft #AIEngineering #Vercel #PortfolioProject #FICKSLABS

Microsoft continuely adds architectural effort building Agentic Frameworks, the new challange is to decouple agents' optimization from execution framework, Agent Lightening can be integrated with any Agent Framework; CrewAI, Semantic Kernel, AutoGen, OpenAI Agent SDK, Azure AI Agent Service, Agent Framework, Langgraph, any Agentic RAG framework r even WITHOUT agent framework (Python OpenAI). By decoupling agent framework from RL training system, Agent Lightning can be seamlessly enables model training for any existing agent, without requiring any modifications to the agent code. It allows model training to be directly aligned with the agent’s real deployment behavior and task logic, resulting in improved performance and adaptability. https://lnkd.in/d2Ws6SCc

Microsoft's New Agent Framework Microsoft Agent Framework is an open-source development kit for building AI agents and multi-agent workflows for .NET and Python. It brings together and extends ideas from the Semantic Kernel  and  AutoGen projects, combining their strengths while adding new capabilities. Beyond merging the two, the Agent Framework introduces workflows that give developers explicit control over multi-agent execution paths, plus a robust state management system for long-running and human-in-the-loop scenarios. The Agent Framework offers two primary categories of capabilities: 1️⃣ AI Agents: individual agents that use LLMs to process user inputs, call tools and MCP servers to perform actions, and generate responses. Agents support model providers including Azure OpenAI, OpenAI, and Azure AI. 2️⃣ Workflows: graph-based workflows that connect multiple agents and functions to perform complex, multi-step tasks. Workflows support type-based routing, nesting, checkpointing, and request/response patterns for human-in-the-loop scenarios. #microsoft #aiagents #agentframework

𝗠𝗶𝗰𝗿𝗼𝘀𝗼𝗳𝘁'𝘀 𝗡𝗲𝘄 𝗔𝗴𝗲𝗻𝘁 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸 Microsoft Agent Framework is an open-source development kit for building AI agents and multi-agent workflows for .NET and Python. It brings together and extends ideas from the Semantic Kernel and AutoGen projects, combining their strengths while adding new capabilities. Beyond merging the two, the Agent Framework introduces workflows that give developers explicit control over multi-agent execution paths, plus a robust state management system for long-running and human-in-the-loop scenarios. Microsoft Agent Framework = AutoGen + Semantic Kernel The Agent Framework offers two primary categories of capabilities: 1️⃣ AI Agents: Individual agents that use LLMs to process user inputs, call tools and MCP servers to perform actions, and generate responses. Agents support model providers including Azure OpenAI, OpenAI, and Azure AI. 2️⃣ Workflows: Graph-based workflows that connect multiple agents and functions to perform complex, multi-step tasks. Workflows support type-based routing, nesting, checkpointing, and request/response patterns for human-in-the-loop scenarios. #Microsoft #AIAgents #AgentFramework

Boost your productivity with 𝗻𝟴𝗻 — the powerful, flexible workflow automation platform that brings AI-native capabilities to your fingertips. Whether you're integrating apps, APIs, or building custom automation with JavaScript and Python, 𝗻𝟴𝗻 offers a no-code/low-code environment designed for developers and teams who want full control without sacrificing ease of use. From automating repetitive tasks to creating AI-driven workflows, 𝗻𝟴𝗻 lets you scale smarter and faster. Try self-hosted or cloud options with over 400 integrations and join a rapidly growing community of innovators transforming automation. #WorkflowAutomation #NoCode #LowCode #AI #AINative #Integration #n8n #AutomationTools #Developers #TechInnovation #DigitalTransformation #ai #artificialintelligence #machinelearning #innovation #technology #aiagents #agenticai #aiagentsystem #aiagentsnetwork #TWM #twmAI #TWMtech #TWMaiagents

Boost your productivity with n8n — the powerful, flexible workflow automation platform that brings AI-native capabilities to your fingertips. Whether you're integrating apps, APIs, or building custom automation with JavaScript and Python, n8n offers a no-code/low-code environment designed for developers and teams who want full control without sacrificing ease of use. From automating repetitive tasks to creating AI-driven workflows, n8n lets you scale smarter and faster. Try self-hosted or cloud options with over 400 integrations and join a rapidly growing community of innovators transforming automation. #WorkflowAutomation #NoCode #LowCode #AI #AINative #Integration #n8n #AutomationTools #Developers #TechInnovation #DigitalTransformation #ai #artificialintelligence #machinelearning #innovation #technology #aiagents #agenticai #aiagentsystem #aiagentsnetwork #TWM #twmAI #TWMtech #TwmAiAgents