This article outlines a method for combining Claude Code and Codex to optimize .NET feature development, overcoming common context and consistency issues in AI tools. The strategy involves Claude for planning and Codex for building, interconnected by a markdown file for efficient communication.
This article discusses an embedded vector database alternative for .NET RAG applications, introducing VectorRAG.Net to reduce latency and eliminate external service dependencies. It analyzes scenarios where this approach outperforms cloud solutions.
This article introduces the Microsoft Agent Framework, positioning it within the current .NET AI stack and explaining its appropriate use cases. It details its importance for agents, sessions, tools, and workflows, building upon `Microsoft.Extensions.AI`.
The article highlights that the main challenge in building Generative AI features in .NET applications using Semantic Kernel is controlling the context sent to the LLM, not merely calling it. Key lessons emphasize creating dedicated context builders to send only relevant data and prioritizing token optimization over debating model versions for better cost and latency.
Phase 2 of the .NET AI Architect Laboratory project enabled an autonomous AI agent, using Microsoft Semantic Kernel, to analyze local project files for security vulnerabilities and performance bottlenecks. The system also implemented dynamic provider routing, allowing it to switch between Gemini 2.5 Flash and Groq (Llama 3.3 / Llama 3.1) on the fly.
This quick tip explains how to implement context compression in .NET for RAG systems, addressing the lack of a direct equivalent to tools like LLMLingua. It proposes using a smaller, cheaper worker model to pre-process retrieved documentation, extracting only essential facts to reduce cost and latency with premium AI models.
A Microsoft anunciou o GitHub Copilot Testing para .NET, uma nova ferramenta de testes de unidade gerados por IA integrada ao Visual Studio 2026. Essa inovação promete um avanço significativo na produtividade de desenvolvedores .NET, automatizando a criação de testes, sugerindo casos de borda e otimizando o processo de testing.
The author rebuilt their TravelAI.Core library to work with multiple AI providers, including OpenAI, Anthropic, Ollama, and Azure, instead of solely relying on Azure OpenAI. This change simplifies the setup process for developers wanting to try the library without an initial commitment to a cloud provider.