Rag and AI Agent for ClicShopping V4, example presentation

[ClicShopping]

This version allows you to generate various types of reports through a conversational interface. We use RAG (Retrieval-Augmented Generation) and an AI agent to enable this functionality.

 

What is RAG?

RAG (Retrieval-Augmented Generation) is an AI technique that combines a language model with a search system. When you ask a question or request a report, the model first retrieves relevant information from a knowledge base or database. Then, it uses that information to generate a precise and context-aware response. This makes reports more accurate and aligned with real-time data.

 

What is an AI Agent?

An AI agent acts as an intelligent assistant that understands your requests and dynamically builds the query or report you need. Based on your input, the agent figures out what kind of data to fetch and how to present it — without you needing to manually define complex filters or SQL.
 

Example Usage:

You simply type your request inside the chat — for example,
"give me  the total stock, the alert stock by categories and products"
The agent understands your intent, retrieves the necessary data, and builds the report automatically using RAG.

below an image what we can do :
 

Development Status

The development is well on the road, and we’ve focused on optimization to make the experience fast and smooth. You just need to type your request, and the system takes care of the rest.

[Fantomas]

Hello @ClicShopping,

 

I do not know a lot about gen ai and what we can do, but what I see it's seems to very interesting.

Thank you for your work

[Kara]

Oh seems to be nice, what I see, we can do some analysis inside the application to understand what's happen in the shop without to create 1000 hundreds statistics dashboard.

This development is nice and original for this king of tools.

[Manupichu]

I have a small shop and Looking more information with chatgpt

The response :
 

Ecommerce Admin Automation

What you can do:

Product Management

Bulk edit titles, tags, prices using smart suggestions or AI prompts.

Auto-generate SEO product descriptions.

Order & Inventory Management

Automate reordering, restocking alerts, vendor communication.

Set up rules like: “If stock < 10 and avg sales > 2/day → notify/purchase.”

Customer Segmentation & Personalization

Group users based on behavior and tailor marketing/messages.

Dynamic homepages or emails depending on the user’s preferences or past purchases.

Fraud Detection

Use analytics and rules to detect unusual buying behavior.

i do not know if there is automation, but it can be useful to have a such functionalities.

[ClicShopping]

Thank you,
The approach for this moment is on 2 orientations :
Semantics : you can look some information as product description, general condition ....
Analytics : Dynamic request allowing to generate report or make analysis.
Lot of work to do.

[Gary]

Hey, this looks promising — nice work! 😁

A few quick thoughts and questions:

Is this feature meant for store admins only, or can it be used by other roles (like warehouse staff or marketing)?

Will it support custom data sources or only internal ClicShopping data?

Can we edit or refine a generated report after it's created? Like filtering or exporting?

Suggestion: It would be cool to see a short video or animated demo of it in action — helps understand how smooth the experience really is.

Also, any plans to allow voice input later on? That could be a fun addition.

Looking forward to testing it when it’s available. Keep it up!

[ClicShopping]

@Gary,

We have decided to start with small featured, and we see in the future how to improve that.
2 orientations, try to give an answer at a question as :
Could you give the description about the SKU-XXXXXXXXX, semantics approach.

And the analytics approach : could you give me the total order amount during the last 3 month?
The last use an agent AI to create dynamically the sql request.