AI-Driven Universal Shopping Portal

I’m building a web-based shopping portal where visitors can ask for literally any kind of product and receive immediate, AI-generated recommendations complete with purchase links. The heart of the platform is an AI chatbot that learns each shopper’s stated preferences in real time, then delivers tailored suggestions—electronics today, groceries tomorrow, a new jacket the next.

Here’s what I need from you:

• A clean, responsive storefront built with a modern stack (React, Vue or similar is fine).
• An integrated chatbot powered by a conversational-AI engine—OpenAI, Dialogflow, Rasa, or another solution you’re comfortable with—that captures user preferences during the chat and feeds them to the recommender.
• A recommendation system that ranks products according to those explicit preferences rather than generic trends or browsing history.
• Automatic generation of outbound product links (affiliate or direct merchant URLs) inside the chat replies and on product pages.
• An admin dashboard where I can add or import product catalogs, manage link formats, and view basic analytics about queries and conversions.

Acceptance criteria:
1. A user can start a conversation, share a few likes/dislikes, and receive at least five relevant product links in under three seconds.
2. The same user can switch to a completely different category mid-chat and still get accurate, preference-aware results.
3. All chat interactions and recommendations log correctly in the dashboard.

If you’ve implemented AI recommenders or e-commerce chatbots before, please highlight what models, frameworks, or APIs you used. Let’s create a seamless, one-stop shopping experience driven entirely by user preferences.