I have implemented practical AI solutions using Azure AI services, including Azure AI Studio for Conversation Language Understanding (CLU) and Azure Machine Learning Studio. Recent work includes developing proof-of-concept applications such as a voice-to-SOAP note generator leveraging custom-trained large language models (LLMs), a CLU-based intent recognition server routing requests to appropriate microservices, and a resume-augmenting chatbot built with .NET 9, Blazor, GPT-4o, and Retrieval-Augmented Generation (RAG) backed by Azure AI Search as the vector database. These implementations integrate AI capabilities into existing .NET/Azure architectures, emphasizing reliable data pipelines, secure deployment in containers/VMs, and observability alongside traditional development practices.

For the record, I think that “vibe coding” for production systems is dangerous. AI-augmented development done by professionals is a force multiplier.  Read my post for complete details.