Over the holidays, I decided to revisit a personal project I worked on in October. Vinyl Scrobbler Recap The project was a very basic Python script prompted using AI. The application sits in the macOS menu bar and allows you to enter a Discogs↗ release ID to scrobble the record you are listening to in real-time to LastFM↗ . The reason for the application was not only to dip my toes into the then newly released Claude 3.5 Sonnet model↗ but also to have the artwork for the album I am listening to show up on my Tuneshine↗ , as existing services didn’t quite have the functionality needed for it to work. ...

This year has been particularly notable for the long-awaited return of Tears For Fears with “Songs For A Nervous Planet” to The Cure’s “Songs Of A Lost World,” we’ve seen legendary artists continuing to push boundaries. The progressive rock scene has been especially vibrant with strong releases from Frost*, Pure Reason Revolution, and The Pineapple Thief. What’s particularly striking about 2024’s releases is the diversity of sounds - from Underworld’s electronic landscapes in “Strawberry Hotel” to Orville Peck’s unique take on country music with “Stampede.” The year has demonstrated that established artists can still innovate while staying true to their core sound. ...

When I last wrote about FLUX.1 from Black Forest Labs, I mentioned, “however, it quickly became clear that my M3 MacBook Pro with its 36GB of RAM wasn’t going to cut it.” Since then, I’ve upgraded to an M4 MacBook Pro with 48GB of RAM. Alongside this, substantial improvements have been made to run FLUX.1 smoothly on macOS machines via the MFLUX project↗ . In this quick post, I’ll walk you through installing and using MFLUX. Let’s dive in! ...

Managing Python environments can sometimes feel like navigating a jungle 🌴🐍, especially when dealing with dependency conflicts. To streamline this process, I created a handy Zsh function to add to your dotfile for switching Conda environments quickly and efficiently. 🚀 Let me walk you through it! Why I Needed This Function 🛠️ After months of using Conda to manage Python environments on my macOS machines, I realized I had built up quite a collection. Switching between them manually became tedious. So, I decided to create a quick, interactive function to simplify the process. 🎯 ...

Like many techies on their day off, I completely abandoned my planned tasks and embarked on an entirely unplanned project: creating a macOS application. Well, to be more precise, I decided to prompt one using Claude 3.5 Sonnet↗ . With Anthropic’s recent updates to their AI model, I figured it was the perfect opportunity to test its capabilities while solving a personal need. The Catalyst: My New Tuneshine Device Recently, I acquired a Tuneshine↗ - a clever device that arrived last week. It works wonderfully with my streaming services of choice, including Apple Music↗ and Spotify↗ , and offers integration with LastFM↗ . ...

If like me you have been guilty of hard coding all of your network configuration in Ansible variables like this … Hard Code Values Lines: 12 Charaters: 359 Language: YAML 1 2 3 4 5 6 7 8 9 10 11 12 vars: network: full_range: "10.0.0.0/24" subnets: - name: "subnet1" address_range: "10.0.0.0/27" - name: "subnet2" address_range: "10.0.0.32/27" - name: "subnet3" address_range: "10.0.0.64/27" - name: "subnet4" address_range: "10.0.0.96/27" ...

As the DevOps landscape evolves, so do our methods for managing tools like Ansible. This post outlines my current preferred approach to installing and managing Ansible on macOS. The Traditional Homebrew Approach For years, many macOS users, including myself, relied on Homebrew to install and maintain Ansible. The process was straightforward: Installing Ansible using Homebrew brew install ansible While this method effectively managed the core Ansible installation, it became problematic as Ansible’s ecosystem grew more complex. ...

This week I decided to see how many buzzwords I could get into one post, given the recent release of Flux by Black Forest Labs I had originally intended on trying to get it up and running locally - however it quickly became clear that my M3 MacBook Pro with its 36GB of RAM wasn’t going to cut it. Because of this, I decided to look at one of the many online services that offer access to the various Flux models via their APIs which led me to Fal.AI which in turn gave me the idea for this post. ...