Help in creating an shopping assistant for retail fashion shop

I am trying to create a shopping assistant for an online retail fashion shop. Our idea is to use chat conversation api to extract entities like category, gender, colour , style, occasion, season etc from the user prompt so that we can send these filters to the retailer catalogue api and display a strip/slider of products in the chatbot . These are some of the examples prompts which we want to action on “Shop for a blue shirt”, “Recommend an outfit for my brothers marriage”, “What should i wear for the job interview”, " Show me a matching shirt for my blue jeans".
We also want to give filter buttons in some cases like for a question from Assistant “Whats your gender” we would like to give buttons for better user experience.
We were able to go halfway using a combination of conversation and function calling results, but it seems its not that much reliable. Could anyone please guide us or is there are any articles we can refer .

Hey there and welcome to the community!

There’s a really good article in the OpenAI cookbook which seems to align with what you’re wanting to do. Have you heard of RAG before? This may help


Thanks Macha , seems like what i was looking for , will see if it fits my usecase by rebuilding the chatbot.

1 Like