These days, I've been gathering information about stocks using AI,
and scraping real-time stock information through LS Securities' websocket API. Based on this, I've been creating reports and various functions.
As a result, I created a function that continuously collects how stock traders or securities experts on YouTube view the current market.
Recently, the articles posted on 주식한당 were written in a format similar to what I wrote after reading them,
and I wondered if I could use the pattern of my writing, which is being run on DGX Spark with qwen 3.6, as a prompt.
Could I not transform recent expert stories into articles like mine by throwing in a specific topic???
So I modified the function twice and ran it...
I made it in a chat format and set it to fetch recent data...
The analysis took about 3 minutes, but the article itself came out quickly.
It does look like the pattern of my writing, but I still feel that AI is a bit limited...
I thought about whether it would be possible to recommend what topics to throw in each time, so I refined the function further.
I was able to implement a function that writes articles using recent content like a daughter.
At this point, it seems possible to create a bot that automatically logs in to Daum and posts automatically...
These days, I've been quietly creating various functions since I bought DGX Spark. It turns out there are a lot of things that can be done with local llm.
I was worried that I might be disappointed before buying it, but the more I use it, the more I realize that I can do so much more. I'm spending my days having fun creating things.
These days, it only takes about 20 minutes to create these simple functions... It makes me think how wonderful it is to live in this world.
