Insights
Practical notes on .NET architecture, Azure/AWS cloud delivery, AI integration, Blazor, Delphi modernization, relational databases, secure APIs, agentic development, and career-focused engineering habits.
Most RAG tutorials recommend a fixed chunk size, such as 512 tokens with 128-token overlap. In production, that approach often breaks business context and reduces retrieval quality. I prefer semantic chunking—splitting documents by logical sections such as policy rules, API endpoints, contract clauses, or code functions. Each chunk should represent one complete business concept.
.NET Architect with 20+ years of experience building scalable, high-performance systems using C#, .NET, and modern cloud-native architectures. Focused on clean architecture, microservices, and pragmatic engineering decisions that balance productivity, maintainability, and performance. Passionate about modern data access patterns (EF Core, Dapper, CQRS) and building systems that scale in real-world production environments.
One question I often hear from developers:"Should I use EF Core or Dapper. "After working on enterprise applications, APIs, and data-intensive systems, my answer is:👉 It depends on your requirements.
