AI assistant for retail and e-commerce
Product information, return policies, promotion terms and store procedures change constantly. Store and call center staff struggle to find the right answer.
▶ The solution
I build an AI assistant that works on your procedures, reports and past projects, answers with sources, and is mathematically optimized for speed and low cost.
▶ Use case
When a store employee asks about a promotion's terms, the assistant answers from the current promotion document and cites it.
▶ Data you need
Product catalogs, promotion documents, return and operations procedures.
The industry's data landscape
In retail and e-commerce, data is spread across point-of-sale systems, the e-commerce platform, loyalty programs and marketplace dashboards. Transaction volumes are high and product ranges wide, so even small improvements in rates add up to a visible difference.
An in-house AI assistant connects large language models (LLMs) to your company's own documents. An employee asks a question; the assistant finds the relevant documents, writes the answer based on them, and shows where it came from.
The math behind it
Monthly cost ≈ N × (t_in × p_in + t_out × p_out)
- N: questions per month
- t_in, t_out: tokens sent to and returned by the model per question
- p_in, p_out: the model's price per token
How I work
- Week 1
Use case and sources.
Who is the assistant for, and for which questions? Which documents are in scope, and which are current? How success will be measured is written down here.
- Weeks 2–3
First working version.
Documents are ingested, the assistant is set up and answers with sources. It is tested with real users' real questions.
- Weeks 4–5
Accuracy and trust.
Answers are measured regularly against a question set. Wrong or incomplete answers are fixed, and the assistant is made to say when it doesn't know.
- Final week
Handover.
How to add new documents, how to monitor the system and how to guide users is documented and handed to your team.
▶ Questions and answers
AI assistant for retail and e-commerce — what data do you need?
Product catalogs, promotion documents, return and operations procedures.
When is this the wrong choice?
You have few documents and everyone knows where they are. A well-organized folder structure and search may be enough. An assistant would be an unnecessary layer. The knowledge is in people's heads, not in documents. No assistant can find knowledge that was never written down. It has to be written down first.
How does this work start?
With a short intro call. I listen to what you need and tell you plainly whether your data is enough for the job and what the first step should be. I reply to every message within 24 hours.
▶ Concepts
Linked to Wikipedia and Wikidata.
▶ Sources
- [1]Wikipedia, Retrieval-augmented generation, 2026-10. en.wikipedia.org/wiki/Retrieval-augmented_generation
- [2]Wikipedia, Large language model, 2026-10. en.wikipedia.org/wiki/Large_language_model
▶ Intro call
Let's talk about your situation.
In a 30-minute call I listen to what you need and tell you plainly how I can help.
Book an intro call▶ In other industries