Words About
Things that interest me.
As someone who generally wants to know what happens under the covers when a button is pushed, I decided to take a different approach to Generative AI. This was the first time I decided I did not want to know how it worked and just wanted to be enamored by how I could use it to enhance my life both in/out of work. I am a software engineer by trade and have a young child so my interests were also how the world will change, which in turn means how different I would have to raise a child. I decided to use Claude by Anthropic as my AI of choice. I have used ChatGPT and found it no as powerful in its capabilities related to code generation. I also wanted to see how a different model would perform, and I was not disappointed at all with my choice. I do think there is a basic learning journey though that will be helpful in getting started that I wanted to share. I follow a four step process which I will detail. Step 1 was to just use the Web UI to and copy/paste into the IDE, Step 2 was to use the in IDE Agent mode, Step 3 was to build on top of that and use Sub Agents and Step 4 which is a work in progress at this time is to create a workflow app with multiple autonomous agents. I would say by the time I got to Step 3 I was really sold on the power that I am able to type a single sentence and get a POC app in minutes, and even get it deployed to my local Kubernetes cluster in short while after that. I will add more along the way but want to give a quick update on my journey so far.
