After looking into Lana’s interview, I realized that besides constantly giving lobster prompts,

we also need to collect data to do ML, so I ran a little bit of data,
and integrated the ML model into the current system.
The benefit of this is that it doesn’t rely entirely on AI. The underlying logic of the code is machine learning.
Then, through Opus 4.7, it performs AI self-adjustment and optimization.
And machine learning is interesting—you can feed it a huge amount of data.
That’s the charm of accumulating time and data.