Background

In today’s digital landscape, services and offerings are increasingly composed of multiple interconnected applications and systems. This complexity makes it difficult to trace which services are tied to a specific customer identifier and to detect issues such as failed orders or recurring system errors.

Traditionally, support and operations teams rely on direct API queries and manual log analysis to access this information. However, this requires technical expertise, slows down troubleshooting, and makes the process less accessible to non-technical staff.

At the same time, conversational AI and the Model Context Protocol (MCP) have emerged as promising technologies to bridge this gap. They enable natural language access to structured (API) and unstructured (log) data, allowing users to obtain insights through simple chatbot-like queries.

High level description

This thesis investigates how AI and MCP can be used to enrich existing APIs with conversational capabilities. The goal is to create a system where users can ask natural language questions such as:

• “Which services are connected to this phone number?”
• “Has this number had any failed orders or logged errors recently?”

The solution should unify access to APIs and application logs through a conversational interface, lowering the technical barrier and improving efficiency in troubleshooting and service insights.

Who are we looking for?

Bachelor/Master of Science in Computer Science/Engineering

Project description

The project will focus on the design, implementation, and evaluation of a proof-of-concept prototype that integrates AI with MCP to provide conversational access to APIs and logs.

Key tasks include:

1. Literature Review:
   • Study related work in conversational AI, retrieval-augmented generation (RAG), MCP integration, and log analysis.
2. System Design:
3. Define an architecture where MCP connects APIs and application logs to an AI assistant.
4. Plan retrieval and integration strategies for structured (API responses) and unstructured (log data) sources.
5. Prototype Implementation:
6. Develop a chatbot interface capable of answering service-related and troubleshooting questions.
7. Connect the prototype to one or more APIs and representative log sources.
8. Evaluation:
9. Conduct experiments and user testing with both technical and non-technical users.
10. Compare conversational queries with traditional API/log access in terms of usability, accuracy, and efficiency.
11. Identify challenges such as reliability, security, and data interpretation.

Purpose and Scope

The purpose of this thesis is to explore how conversational AI and MCP can improve accessibility and efficiency in accessing service and error data across complex digital ecosystems.

The scope includes:

• Designing and prototyping a conversational interface for service lookup and error detection.
• Exploring the integration of MCP with APIs and logs.
• Evaluating the usability and feasibility of the approach in practice.

The work is limited to a proof-of-concept prototype and does not aim to deliver a production-ready system. However, the results should provide valuable insights and a framework that can guide further development and potential industrial adoption.

An Exciting Journey with Knightec Group
Semcon and Knightec have joined forces as Knightec Group. Together, we are Northern Europe’s leading strategic partner in product and digital service development. With a unique combination of cross-functional expertise and a holistic business understanding, we help our clients realize their strategies – from idea to complete solution.

Practical Information
This is a thesis position, located at our office in Sundsvall. Start date January or March 2026.

Please submit your application as soon as possible, but no later than 2025-11-30. If you have any questions, you are welcome to contact Johanna Edström. Note that due to GDPR, we only accept applications through our careers page.