MASTER THESIS AI-Driven Enterprise Architecture Assistant

At Sandvik Coromant were at the forefront of manufacturing tools and machining solutions. Our leadership extends beyond hardware; were driving a digital transformation to define the future of the industry.

Were looking for a Masters Thesis student to join our Enterprise Architecture (EA) team for 20 weeks. Your mission is to explore how Artificial Intelligence can democratize architectural knowledge enabling us to support our global organization faster smarter and with greater precision.

Background and scope of the project

As our digital landscape grows so does the demand for architectural guidance. We need to move beyond static documentation to a dynamic AI-supported service model.

We seek to evaluate and implement an intelligent agent capable of serving the organization as a first line of support answering FAQs clarifying technical guardrails and guiding engineers through complex standards.

Work description

In this project youll explore how enterprise needs and AI techniques can meet in practical real-world ways. Rather than aiming for a quick prototype youll develop a wellreasoned concept that strengthens your thesis work and builds your understanding of architecture and applied AI.

1. Demand discovery and use case definition:

• Conduct intake interviews with engineers and stakeholders across the organization.
• Identify high-value friction points (e.g. specific recurring questions on cloud standards API patterns or security compliance).
• Define the Customer Journey for an employee interacting with the EA team.

2. Technical evaluation & strategy:

• Hypothesis: While tools like Microsoft Copilot Studio offer a low-code baseline your task is to critically evaluate if this is the optimal solution.
• Landscape: Investigate and compare technical architectures including Custom RAG (Retrieval-Augmented Generation) pipelines Open Source LLMs or Agentic workflows.
• The decision framework: Produce a trade-off analysis based on:
  • Architectural Integrity: Can the agent reason correctly about our specific IT ontology
  • Data Sovereignty: Does the solution strictly adhere to our security and privacy standards
  • Maintainability: How easily can the agent learn new facts without complex code deployments
  • Cost vs. Performance: A pragmatic analysis of token consumption and latency at scale.

3. Pilot realization & continuous learning

• Develop a functional MVP (Minimum Viable Product) based on your winning architecture.
• Implement a feedback loop mechanism where the agent improves based on user interactions.
• Deliver a roadmap for operational maintenance and future scaling.

Student background

Were looking for a student whos genuinely interested in Applied AI and curious about how it can be used to solve real challenges in an industrial setting.

• You are studying for a Master of Science in Computer Science Software Engineering AI/ML Information Systems or a similar field.
• You have a basic understanding of Large Language Models (LLMs) prompt design and how context windows affect model performance.
• Youre comfortable working in languages like Python C# or similar.
• You take an analytical and objective approach you dont choose tools because theyre trendy but because they make sense for the problem youre solving.
• You can communicate clearly and translate complex technical ideas into something others can understand whether theyre engineers or business stakeholders.

Why Sandvik Coromant

• Impact: Your work will directly influence how our global engineering community accesses information.
• Innovation: Work in an environment where Manufacturing Wellness and digital sustainability are core values.
• Culture: We value passion fair play and an open mind. You will be fully integrated into our team with mentorship from the Head of Enterprise Architecture.

We believe that diversity of experience perspective and background leads to a better environment for our employees our business and our customers. Now were looking forward to meeting you!

Duration and location

Location: Sandvik Coromant HQ (Sandviken) or Flexible/Hybrid

Scope: 30 Credits (20 Weeks)

Contact information

For more information about the thesis program please contact:

Emil Ciepielewski Head of Enterprise Architecture at

Application

Please send in your application as soon as possible as selections are made on an ongoing basis.

Sandvik Coromant is part of the global industrial engineering group Sandvik and is at the forefront of developing manufacturing tools and machining solutions with knowledge that drives the industry standards and innovations demanded by the metalworking industry now and in the next industrial era. Collaborations with educational institutions extensive investment in research and development and strong customer partnership ensure the development of machining technologies that change lead and drive the future of manufacturing. Sandvik Coromant owns over 1700 patents worldwide employs around 8000 staff and is represented in more than 150 countries.