Doctoral Student in Intentional Monitoring of Agentic AI Software

About the role

Four-year fully funded PhD at Chalmers in Gothenburg, focused on intentional monitoring of agentic AI systems. The work develops runtime guards, intent recognisers, and audit trails that make autonomous LLM-driven software inspectable and accountable. Sits at the intersection of software engineering and AI safety. Deadline 12 June 2026.

Responsibilities

• Design observability and monitoring stacks for LLM-driven autonomous software.
• Build intent-recognition layers that infer and log agent goals at each step.
• Validate the methods on real agentic systems from industry partners.
• Publish in ICSE, FSE, ASE, and AI-safety venues.

Requirements

• Master’s degree in computer science, software engineering, or a closely related field.
• Solid programming skills in Python and at least one systems language.
• Familiarity with LLM tool-use frameworks (LangChain, AutoGen, or similar).
• Fluent English. Swedish is not required.

Nice to have

• Prior research on AI agents, program analysis, or runtime verification.
• Background in formal methods or software testing.
• A first-author publication or substantial open-source contribution.

How to apply

Open the apply link and submit CV, transcripts, motivation letter, Master’s thesis, and two reference contacts via the Chalmers recruitment system before 12 June 2026.