Would you consider open-sourcing the client-side code? For privacy-focused PDF tools, that seems like the easiest way to make the “no upload” claim more trustworthy.
The output head byte budget is surprising. Did you try any tradeoff where the head is compressed more aggressively but experts stay mostly untouched?
Interesting concept. How do you make sure the “navigator” helps the user understand the reasoning instead of becoming another layer of AI-generated abstraction?
Would you consider open-sourcing the client-side code? For privacy-focused PDF tools, that seems like the easiest way to make the “no upload” claim more trustworthy.
The output head byte budget is surprising. Did you try any tradeoff where the head is compressed more aggressively but experts stay mostly untouched?
Interesting concept. How do you make sure the “navigator” helps the user understand the reasoning instead of becoming another layer of AI-generated abstraction?