Project DescriptionThird-cycle subject: Computer Science This doctoral position is part of an ambitious 5-year Wallenberg Scholar project titled “Scalable and adaptive inferencing for democratizing AI”. With a project budget of 18 million SEK, the goal is to dramatically reduce the cost and power required for serving large language models such as ChatGPT. Early publications and project details are available at: https://www.kth.se/blogs/dai/ The work will primarily involve:
Supervision: Professor Dejan Kostic and Associate Professor Marco Chiesa. Final admission decisions will determine supervision. What We Offer
Admission RequirementsApplicants must meet basic eligibility under Chapter 7, 39 § of the Swedish Higher Education Ordinance:
Additionally, English proficiency equivalent to English B/6 is required. Selection CriteriaSuccessful candidates should demonstrate:
Preferred background: Computer Science, Engineering, Data Science, or Machine Learning. Knowledge of systems and networking is highly desirable. Personal qualities: Scientific curiosity, goal-orientation, perseverance, and strong problem-solving abilities. Target DegreeDoctoral degree Information Regarding Admission and Employment
Union RepresentativesInformation on union representatives and doctoral section contact details are available upon request Application InstructionsSubmit your application via KTH’s recruitment system, including:
Deadline: Applications must be received by 15 January 2026, by midnight CET/CEST.
We firmly decline all contact from recruitment agencies or marketing personnel. Join KTHKTH is a leading international technical university, fostering creativity, research, and innovation to advance sustainable society. Employees benefit from good working conditions, equality, diversity, and professional growth opportunities. Learn more about working at KTH Type of Employment
Location: Stockholm, Sweden Contact
|



















