We have already build a chatbot for answering FAQs and it is working pretty well and we are still trying to more functionalities using custom actions.Apart from this we are looking forward to build a chatbot in my native language like Marathi , Hindi, etc. As far as I know we need to use tensor flow embedding for it. What are other things we need to do ?
I would request Rasa team to share some tutorial regarding this which definitely help many developers around the world to develop the chatbot in their native languages.
I am flattered a bit knowing you believe i am working for Rasa, however I am just a fan and an active community member
Anyhow, one of my first blog was about the tensorflow embedding and i used it to test out a bot in Bengali using latin alphabets.
I also stiched another one (taking inspiration from pre-written github answers regarding the usage of FastText and spaCy)
Both shows how to build a chatbot using non-traditional languages.
Keep in mind, they are written with the previous versions in mind but this might give you an idea of what are the different steps. Tensorflow model will require some good amount of data since you are training embeddings from scratch.