Working with Generative AI is cool. As a user, you mostly see the chatbot as frontend of an AI application with amazing results produced by Large Language Models (LLMs). However, as a developer, I need more insights during the development phase. Specifically, I want to know what information is being passed to my LLM. That’s why I usually enable tracing for my AI applications, but how can I do this in Prompt flow? That is not so difficult. So let…
Custom Connectors and APIs are a pleasure for Power Apps. I can access my needed data, the setup is very fast, and finally I can use the methods of my connector from my Power App. This works fine, except when my API call fails. How can I handle such errors in my PowerApps? Is there something like a Try-Catch-Block? To be honest, it took me a while to find a setup that works for me. So let me first explain…
Recently, I realized that there is a very simple way to prepare and share telemetry reports. This are Jupyter Notebooks. A Jupyter notebook is ideal for combining documentation texts and telemetry queries. This means that my explanations and recommendations for action are part of my telemetry reports. I like this possibility. This helps us at COSMO CONSULT to create prepared reports for monitoring and analysis of Microsoft Dynamics 365 Business Central in a simple way. Prerequisites Azure Data Studio Python…
Any kind of automation will lead into losing information about the magic in the background. You see only the results like “Build Succeeded” or “Build Failed”. That’s fine if you manage one or two Azure DevOps build pipelines, but this is terrible if you maintain hundreds or thousands! What happens is predictable. If you don’t know what is going on in the background and what causes problems – you are in trouble. You have to investigate issues direct in each…