A blog about things I learn at the keyboard: DevOps, CI/CD, Cloud, Automation, to name a few...
A natural-language analyst that connects evidence across the data landscape, routes questions to governed models, applies the required filters, runs queries under your identity, and shows its work. The LLM handles the conversation; governed models and model-specific guidance determine how each question is answered.
How I wired together GitHub Copilot, Work IQ MCP, and Azure DevOps MCP to turn meeting transcripts into draft work items I review and approve.
How I set up Squad, an open-source AI agent team framework, and what I learned along the way
Shipping an AI Agent MVP: What Worked
Optimizing GitHub Actions Workflows for Speed and Efficiency
I built a VS Code extension to toggle VPN connections because clicking is too much work
TL; DR: I created a VS Code extension that makes it easier to remove credentials from Git History.
In 2020, I blogged about associating automated tests with Azure Test Cases. The post had 18 questions, which indicates there is still confusion on how this works, especially how to set it up with Playwright (which was pre-stable release at the time).
I was recently tasked with measuring the impact of a Redis cache on an e-commerce site. This was pretty simple with Azure Load Testing, by comparing the results of 2 sites, one with cache, and one without. However, to better exercise the site and understand the user experience, I wanted also to use Playwright.
I’ve been having a lot of fun with GitHub Actions lately and wanted to document some of the features I regularly use, including some tips and tricks.