Master Thesis – Strategic Approach for Agentic AI Adoption in Sourcing

Axis Communications has tested an AI agent for supplier quality audits and now wants to explore its wider potential within Sourcing. This thesis will investigate where agentic AI could support sourcing processes, what gaps and risks must be addressed, and when simpler automation would be a better choice. The aim is to recommend a prioritized, feasible approach to future adoption rather than develop production-ready software or promise efficiency improvements.

The assignment

• Review sourcing workflows and identify repetitive, data-intensive activities that could benefit from AI agents.
• Assess gaps in data, process documentation, skills, infrastructure, APIs and policies that could affect reliable implementation.
• Develop a framework for prioritizing use cases and compare agentic solutions with alternatives such as workflow automation, search and conventional AI assistants.
• Evaluate risks and constraints, including hallucinations in important decisions, data privacy and dependence on large language models.
• Outline a conceptual multi-agent structure with chains and handoffs for end-to-end sourcing processes.
• Combine a literature review, stakeholder interviews and process mapping, refining the research methods and evaluation criteria with the supervisors.
• Produce a strategy and feasibility report covering prioritized opportunities, required actions, potential efficiency gains and recommended next steps.

Who we are looking for

• Master’s students in Computer Science, Industrial Engineering or a related discipline, particularly with a focus on digital transformation, supply chain management, or AI and data science.
• Strong interest in combining AI with business process improvement, supported by an understanding of LLMs and how AI agents differ from ordinary chatbots.
• Strong analytical and documentation skills, with confidence interviewing stakeholders and connecting business needs to technical possibilities.
• Knowledge of sourcing or supply chain processes, prompt engineering, AI agent frameworks or BPMN is advantageous.
• Applicants must be affiliated with a Swedish university or university college, either directly or through an exchange programme.

What we offer

• Supervision from Axis contacts in GSQ and CM, with the research approach developed together with the students and their academic supervisor.
• An opportunity to examine practical AI adoption challenges in a global supply chain context.
• An open, friendly and collaborative working culture that encourages learning and new ideas.

Practical details

Location: Lund. Spring 2027 start, with flexibility on the start date. The thesis is carried out by two students and comprises 30 credits per student. Apply through the posting by 20 November 2026, in Swedish or English, attaching a CV and university grade summary. Apply as a pair: each student submits a separate application and names their co-applicant. Supervisors and contacts: Christian Fagerlind (GSQ) and Axel Mauritzson (CM).