sam131725/public_grievenes ? reverse-engineered prompt
Reverse engineered prompt
Build me a multilingual public grievance system that can read citizen complaints in English, Hindi, Marathi, and other similar languages, then figure out which department it belongs to, how urgent it is, and whether the message sounds negative, neutral, or positive. I want it to clean and prepare complaint data from CSV files, handle missing or messy fields, and keep track of whether labels came from real data or were filled in by simple rules. It should also show how well the model is doing overall and for each language, including a confusion matrix and a list of wrong predictions so we can review mistakes.
Please make it usable two ways, through a simple API where I can send one complaint or a batch of complaints, and through a small dashboard where I can paste text or upload a CSV. It would be great if there is a fast baseline model and a stronger multilingual model option, with results saved so I can compare them easily.
Are you gonna build this?
make sure you review the code using coderabbit