Semantic Kernel

The latest news from the Semantic Kernel team for developers

Semantic Kernel’s Ignite release: Beta8 for the .NET SDK

Yesterday’s and today’s  Ignite sessions have been jam packed with updates. From new AI infrastructure to the new Copilot Studio, Microsoft has had a lot to share! At the center of it all, however, is still AI Orchestration: the layer that uses AI to retrieve data, call functions, and ultimately do real work. That’s where Semantic ...

Introducing v1.0.0 Beta6 for the .NET Semantic Kernel SDK

The Semantic Kernel team continues to make improvements to our beta release of the .NET library. Our sixth Beta release is the most jam packed one so far, so we wanted to take the opportunity to share everything new inside of it. As a reminder, if you want visibility into our full plans for v1 of Semantic Kernel, please check out this blog ...

Building AI Applications with Microsoft Semantic Kernel and MongoDB Atlas Vector Search

As part of this guest blogpost by the MongoDB Atlas Vector Search team, we are excited to announce native support for MongoDB Atlas Vector Search in Microsoft Semantic Kernel. With this integration, users can bring the power of LLMs (large language models) to their proprietary data securely, and build generative AI applications using RAG (...

Unveiling the Next Steps in AI Productivity: Plugins, Planners, and Personas at Microsoft Ignite

As the Microsoft Ignite event approaches on November 16th, we are excited to invite you to join us for a live session at the Seattle Convention Center Summit. This gathering is a special opportunity for business leaders and AI enthusiasts to gain insight into the advancements of Semantic Kernel, particularly in the realms of Plugins, Planners...

OpenAI Assistants: the future of Semantic Kernel

During the OpenAI event earlier today, OpenAI announced the launch of GPTs and the assistants API, the new and improved ways of creating agents on top of their chat completion models. With assistants, much of the heavy lifting required to build agents has been stripped away… This ultimately means it’ll be faster, and ...

AutoGen Agents Meet Semantic Kernel

In this blog post, we show you how you can use Semantic Kernel with AutoGen, a Microsoft Research project that shows the potential of using multiple agents together. With AutoGen, Microsoft research has shown that multiple agents can be better than one. To actually do work, however, agents within AutoGen need hooks into the real world so...

What to expect from v1 and beyond for Semantic Kernel.

(image) In a previous article, we announced the beta launch of Semantic Kernel v1. In that article, we shared the initial breaking changes we made for v1: 1) renaming skills to plugins, 2) making Semantic Kernel AI service agnostic while still supporting first class support for Azure OpenAI and OpenAI, and 3) consolidating our implementation...