What I Did Today: Learning, Work, and My Homelab
TypeScript, SQL, GPU plans, and architecture quanta
AI helped me with the grammar in this post.
This morning, I went through some basic TypeScript rules that I had forgotten. I also went through SQL syntax to refresh my memory.
Then I explored Vast.ai, looking at its architecture and pricing. I was thinking about renting a GPU to test different models instead of buying a new PC to run those models locally. This could be my first step to get more familiar with the topic and later make an informed decision about which PC and GPU to buy.
Buying a GPU might not pay off in the short term. I am also not sure how much I would use it for my own projects. So I postponed the decision while improving my understanding of the topic.
At work, I was reconciling changes from an external team. This took almost my whole day, so I had to put off working on a new feature. I want to spend less time reconciling changes and more time working on features and thinking about the project. I should probably set a rule: collect changes across the repositories and spend a few hours on one day each week reconciling them.
I also bought 16 GB of RAM for my homelab PC, and I’m happy with that. Now the memory usage is around 16%, so I have plenty of room for new applications.

I also studied “Scope of Architecture Characteristics” from the book Fundamentals of Software Architecture. I learned about architecture quanta. A quantum is an independently deployable part of a system, with its parts working toward one purpose. It includes the dependencies it needs to work, such as a database. In the example below, the service waits for the database, so a slow database also means a slow response from the service.
