@sarit maitra
This is part 2 of 2nd question. I kept the front-end work in a separate repository to avoid cluttering in one repo.
Here I have created end point API apps using FLASK and FAST API.
Both the apps are tested with a small test data enclosed to the programs. However, I do not have a predictive model yet. The apps are based on data modeling and named entity recognition. I have created FAST API which has swagger tool to test the app.
Below screen shot is from Swagger tool.
Creating an API end point won't be that of a complex task provided we have the the necessary data structure available.