As our new age of agentic AI takes shape, Google is stepping up its game with the introduction of the Jules agent and enhanced Gemini AI, which could be seen as a direct response to the growing competition from Microsoft and GitHub, who have been making significant strides in the space.
Microsoft ushers in a new era at Build 2025, unveiling a wave of AI-powered agents designed to transform software development, supercharge productivity, and redefine the future of coding.
OpenAI's "Codex" AI model is back, in a new form from the 2021 offering that powered the original GitHub Copilot and kickstarted the GenAI craze.
Microsoft this week shipped the fourth preview of .NET 10 without groundbreaking features but with much work devoted to ASP.NET Core & Blazor, where several work items were devoted to OpenAPI.
Agent mode, now in public preview for Visual Studio 17.14, marks a major step forward for AI-assisted development. Unlike previous Copilot features, agent mode can autonomously plan, edit, iterate, and invoke trusted tools-completing complex coding tasks from a single natural language prompt.
Microsoft, contradicting beliefs of Java developers responding to a survey, said they don't need to learn AI, master machine learning, or switch to Python to build intelligent, production-ready applications.
The new way to get the most out of GitHub Copilot is from markdown prompting, or writing detailed, reusable natural-language instructions in files like README.md or copilot-instructions.md to guide different AI models in generating context-aware, accurate code.
Uno Platform, a .NET-centric open source project for building single-codebase apps across multiple platforms, this week announced v6.0 of its flagship offering, which introduces a zero-install, web-based IDE for rapid cross-platform development, alongside a modernized app architecture that embraces MVU and .NET-style extensions.
"What MCP server would be helpful to use with Copilot in Visual Studio? I want to write one."
Cursor and GitHub Copilot go head-to-head in a pair of firsthand reviews. One coder returns to Copilot after it adds support for top LLMs. A coding writer falls for Cursor’s conversational style and beginner-friendly flow.
A new GitHub repo serves as a lab for working with the Model Context Protocol (MCP) in Microsoft Copilot Studio.
Andrej Karpathy, who started the whole "vibe coding" thing to describe AI-drive software development, is now warning developers to keep tight control of their "new over-eager junior intern savant with encyclopedic knowledge of software, but who also bull* you all the time, has an over-abundance of courage and shows little to no taste for good code."
Microsoft announced the availability of .NET Aspire 9.2 and the second preview of the .NET AI Chat Web App template, highlighting new capabilities that bring cloud-native orchestration and AI integration closer together in the .NET ecosystem.
Preview 3 features next edit suggestions -- or tab, tab, tab coding -- following previous previews of GPT-4o code completions and assorted Copilot enhancements.
On the way to autonomous AI, Microsoft announced an early access research preview of "computer use" for Copilot Studio wherein AI agents visually interact with any app or website -- clicking, typing, and navigating like a human.
dtSearch, a specialist in enterprise and developer text retrieval software and document filters, announced it's now powering an upgrade to KaleidoSearch from Contegra Systems.
While Preview 3 comes with a wide array of incremental improvements across performance, libraries, CLI tooling, ASP.NET Core/Blazor and more, it doesn't introduce any major new features.
The Model Context Protocol (MCP) for agentic AI has gained much traction since being introduced by Anthropic last November, and now it has a C# SDK.
Microsoft announced agent debugging functionality for Microsoft 365 Copilot directly from the AI tool itself, no Visual Studio 2022 or Visual Studio Code needed.
GitHub chose Microsoft's 50th anniversary to highlight a bevy of Copilot enhancements that further the practice of "vibe coding," where AI does all the drudgery according to human supervision.