My Strategy on AI

AI is all over the place, everyone talking about it and there is a lot of noise. As someone who is against the hype, I have to say that we will keep hearing about AI and its impacts for a long time.

LLM's are not that bad! Most of the time they're wrong, but they can be really helpful, especially for productivity. So my plan is to use them more and more, apply them to my workflows and hopefully make my life easier. But I'm totally aware about their issues with privacy for our data. My goal is to build solutions that are privacy focused and locally runnable. I know running LLM's locally will require lots of computing power for complicated tasks, so I may need to find a way to leverage the power of cloud hosted LLM's without sacrificing the privacy.

Short term goals

Use what is available and works for my projects and work tasks. If my tasks are super repeating, I probably will write a script to make it faster and easier. For example I'm building a cli tool to use LLM's on my terminal and terminal outputs.

Mid term goals

I'm looking to build AI agents that can do most of my daily work, so I can focus on the fun parts :^) .

Long-term goals

Make money with AI agents working for me. This can be possible by building swarm of specialized agents that are good at only one thing. They can find a problem, build a solution and iteare over it without my help. If this happens and I don't publish the result, you can check it by looking at my GitHub bio (just like dylan: https://github.com/dylanaraps).