🚀 Introduction​

I'm planning on setting up a k3s cluster and Windows Server environment in my homelab, which means I need to provision quite a few VMs. Clicking through installers definitely isn't very "as code" 😅, so I've turned to Packer to build VM templates with the idea that I can prepare VM templates with remote management enabled by default.

Now Ubuntu offers Ubuntu Cloud Images which can easily converted into Cloud-Init Templates for use with Proxmox. But but since I'll need custom images for Windows anyway I figured I'd build my own Ubuntu templates too. Its a good opportunity to get some more experience with Packer and gives me full control over the template creation process.

In this post, I’m sharing my plans for my homelab in 2025. I’ve been running a homelab in some form for years now. It’s been a great way to learn, experiment, and stay hands-on with new technologies. As we head into 2025, I’m taking a fresh look at my setup with some clear goals in mind. Here's what I’m aiming to build, explore, and improve in the coming year.

As cloud engineers and developers working with Azure, we've all been there - forgotten resources silently accumulating costs in our sandbox environments. After one too many surprise bills, I decided to build a solution that would automatically clean up unused resources in Azure subscriptions.

There are many tools and scripts available for to accomplish this, But in this blog post i'll share the solution im using to keep my Azure costs under control.

A number of the workloads I run in Azure are based around Python Function Apps. I thought it would be a good idea to create a Terraform module that can be reused across multiple projects while leveraging out some of the newer features in Terraform: Terraform Test framework and Write-only arguments.

View the Module on GitHub

In this post, we’ll explore how to replicate GitHub Copilot’s pull request (PR) description functionality within Azure DevOps, using a custom solution I’ve called DescriptoBot. The goal is to automate the creation of PR descriptions by leveraging Azure OpenAI models, providing a streamlined, AI-driven method to summarize changes and improve efficiency across teams.

In this post, we’ll explore how to configure Git to use different usernames based on the directory you’re working in. This can be useful when working on multiple projects with different email addresses associated with each one. eg personal projects, work projects, client projects, etc.