Getting Ready for WWDC And Building my HomeLab

It’s been an eventful year, and WWDC is right around the corner (tomorrow as I write this). every year since 2018, I’ve taken an Education-Vacation to align with WWDC. In 2018 and 2019, I was lucky enough to actually get a ticket to go in person. I’ve tried the last few years, but no luck so far.

Picture of author at the WWDC conference in 2019.  It's a selfie with the glowing logo of WWDC19 over the left shoulder, and a presentation stand.

WWDC 2019 was amazing, and I had a great time meeting other developers and talking to Apple engineers. It was also the year that the first developer build bricked my MacBook Pro, and I had to schedule time to have the Apple team to re-image my machine as if the Beta came from the factory! What an experience that was.

Well this year, I will be squirreled away at home for 8-12 hours a day, going thru videos and setting up my development devices to try and take advantage of the new APIs and features that Apple will be releasing tomorrow. I hope that they will make my Apps more competitive in the App Store.

In the mean time, I have been doing a lot more work with AI models at my day job. I’ve been working for a year at Atlassian, as a Principal in their Advisory Innovation Lab. My job is to help build those solutions that our consulting teams will use with large enterprise customers. I am primarily focused on AI governance, which is a huge and important things for enterprises right now. And as such, I’ve been spending a lot of time doing some amazing things to help build out systems and processes for the safe, secure, and appropriate usage of AI tools.


This approach helped me focus on setting up my own private AI stack at home, and over the last few weeks, I’ve been building a set of processes so that I can ask questions about my code and gain insights to patterns that have been alluding me.



Today, I have finally finished a web-based chat setup using Ollama, LM Studio, Open WebUI, Qdrant, and custom python code. I can now load up a chat with any of my major projects, it automatically indexes any code changes to update the RAG model I have setup, and I use qwen2.5-coder-32b-instruct with a large context window.

The one challenge I am still working on is how I can take advantage of a great set of Swift skills developed by Paul Hudson over at hackingwithswift.com. Once I can figure that part, I am hoping to add Apple’s latest Swift developer documentation to my RAG in order to ensure that I am always using the latest APIs in my discussions.


Like my good friend Andy Piper, I don’t like the idea of my code or data being consumed by large enterprises for their advantage. If they want to pay for, I am sure we can come up with a financial arrangement, but until then, it is my code and my ideas.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.