October is ending soon, and I only have two months left for this year. I had three major goals for this year, but like what I said earlier this year, they are more like guidelines. For sure, I have worked on them, but the outcome doesn’t match my expectations. Well, it isn’t a bad thing, I guess.
Currently, I am working on a few things.
One is the wowsinfo-core project and its MVP. This one is just about using Kotlin Multiplatform as a future migration from my existing React Native, so I can still keep the React Native part unchanged, but start using the shared common logic including service, usecase, repo and models. The main project will just keep going when I feel like it, there is no rush.
The other one is about LLM finetuning using LoRA or something else. This is the extension of my conversion project, but not sure what I will get from it. I really want to see how long this will take and if the model can truly gain additional knowledge the way I except. My data preparation is almost done, so I may start the finetuning soon.
Then, I can probably continue with my trading system, not the model. As I mentioned before, I have to move forward with a working system whether it is profitable or not. I have to see how it goes in the paper trading, instead of my local backtesting to know all the real problems. This upcoming Christmas, I will have 3 weeks off, so I will definitely do something about it after resting for a few days.
There are other things I may do. After working for 4 years, I feel like things are finally stabilising, hopefully. That’s why I may start to work more on my life other than all these projects. My friends are getting married and having kids, and I am still here. Well, we all have different paths, but it could be the time for me to start thinking.
Just in about a month, it will be the JLPT again once a year, busy but always fun to see everyone again every year. See, it isn’t too bad, right?