Design and Implementation of an AI Agent : customer support in Finnish SMEs
Raza, Azeem (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025060921855
https://urn.fi/URN:NBN:fi:amk-2025060921855
Tiivistelmä
In today’s fast-moving business world, customers expect fast, personalized, and multilingual support as a normal part of good service. Big companies often have the tools and budget to use advanced AI to meet these needs. But many small and medium-sized businesses (SMEs) in Finland struggle to find options that are both affordable and flexible especially for handling more than one language. This thesis looks into how to design and build an AI customer support agent made for mobile retail businesses, focusing on the needs of Finnish SMEs.
The system uses the advanced LLaMA3-70B language model to understand and respond to questions in both Finnish and English. It can figure out the purpose of a message whether it is about a product or an order and give useful replies in real time. Tools like LangChain and LangGraph help manage the conversation flow, and the Groq API keeps responses quick. A simple, user-friendly interface was created with Streamlit to make it easy to use. Tests showed that the system helps businesses reply faster and keeps customers happier. While this project focuses on Finnish SMEs, the system can be adapted for other industries and languages, showing how AI can really improve customer service.
The system uses the advanced LLaMA3-70B language model to understand and respond to questions in both Finnish and English. It can figure out the purpose of a message whether it is about a product or an order and give useful replies in real time. Tools like LangChain and LangGraph help manage the conversation flow, and the Groq API keeps responses quick. A simple, user-friendly interface was created with Streamlit to make it easy to use. Tests showed that the system helps businesses reply faster and keeps customers happier. While this project focuses on Finnish SMEs, the system can be adapted for other industries and languages, showing how AI can really improve customer service.