Extracting specific data from pdf - fine tuning

Hello everyone, i hope you’re doing well
I use this approach to extract data from pdf using the tool file_search :
1 _ Uploading the file
2_ Creating Thread
3_ Sending Message
4_ Create a run and Poll for the response (using an assistant that i already created)

** And I want to know is this the best approach, and If i can Fine-tune a model to do better for this case. **
If so, please provide me with some guidance