The University of Agder (UiA) invites applications for a PhD Research Fellow in ICT, focusing on Multimodal AI for Fast and Accurate Investigation and Response. This full-time position is based at Campus Grimstad and is affiliated with the Department of Information and Communication Technology within the Faculty of Engineering and Science. The role is directly linked to the Research Council of Norway Innovation Project, F AI R, where UiA is a core research partner. The successful candidate will join the Centre for Artificial Intelligence Research (CAIR) and the Centre for Integrated Emergency Management (CIEM), contributing to advanced research at the intersection of artificial intelligence, investigation, and emergency management.

The research will be conducted in realistic, data-intensive environments, addressing challenges related to robustness, reliability, evaluation, and trust in AI systems deployed in high-stakes police operational settings. The project involves close collaboration with industry and research partners, including DavidHorn, and emphasizes applied research in multimodal and temporally structured data. Key research themes include multimodal machine learning, representation learning, embedding-based methods for information extraction and similarity search, evaluation and benchmarking of complex AI systems, and trustworthy AI. The candidate will explore neural architectures for robust and aligned embeddings across text, audio, video, and time-series data, supporting retrieval, information integration, and decision support in evolving environments.

Methodologies will focus on rigorous evaluation and benchmarking beyond conventional text-only benchmarks, including robustness, uncertainty, temporal consistency, and performance under distribution shift. Explainability and interpretability methods tailored to complex multimodal and embedding-based systems are central, ensuring transparency and inspectability for expert users. The research approach combines theory, methods, and empirical experimentation, validated through controlled experiments and realistic pilots in data-intensive environments. The candidate will contribute to high-quality scientific publications and reusable benchmarking frameworks, promoting open and reproducible research practices within CAIR and CIEM.

Applicants must hold or be near completion of a Master’s degree in Computer Science, ICT, Artificial Intelligence, or a closely related field. Essential qualifications include a solid foundation in AI, machine learning, or computer science, strong programming skills (especially Python), and proficiency in English. International candidates must meet English language requirements (TOEFL or IELTS). Desired qualifications include experience or strong interest in multimodal AI, large language models, embedding-based methods, explainable and trustworthy AI, and collaborative development workflows. Familiarity with evaluation and benchmarking of AI systems, robustness, uncertainty estimation, bias analysis, and failure-mode analysis is preferred. Prior research experience and interest in open and reproducible research practices are advantageous. Personal qualities such as motivation, independence, analytical skills, communication, teamwork, inventiveness, and willingness to engage in an international research environment are important.

The position offers professional development in a large, socially influential organization, a positive and inclusive working environment, modern facilities, and membership in the Norwegian Public Service Pension Fund. The gross annual salary is NOK 550,800, with a compulsory pension contribution. UiA encourages qualified candidates from diverse backgrounds to apply. The application deadline is 24 February 2026. For questions about the position, contact Professor Morten Goodwin, Professor Jaziar Radianti, or Head of Department Folke Haugland. For application process inquiries, contact HR-advisor Linda H. Kristiansen. Applications must be submitted electronically, including certificates, Master’s thesis, references, academic work, project description, and other relevant documentation. All documents should be in a Scandinavian language or English.