This PhD position at KTH Royal Institute of Technology is part of a prestigious, five-year Wallenberg Scholar project titled "Scalable and adaptive inferencing for democratizing AI." The project, with a budget of 18 million SEK, aims to dramatically reduce the cost and energy required to serve large language models such as ChatGPT. The research will focus on the design, implementation, and evaluation of distributed systems and networks for machine learning inference, as well as the application of machine learning concepts to develop agent-based frameworks. The successful candidate will join a dynamic, international research environment, collaborating with industry and leading universities worldwide. The position offers full-time employment with a monthly salary according to KTH's doctoral student salary agreement, along with a range of employment benefits. The doctoral studies are expected to be completed within four years, with the possibility of limited teaching or administrative duties (up to 20%). Supervision will be provided by Professor Dejan Kostic and Universitetslektor Marco Chiesa, both recognized experts in the field. The research will be conducted within the School of Electrical Engineering and Computer Science at KTH, in Stockholm, Sweden. Eligibility: Applicants must have a master's degree or equivalent in a relevant field, or have completed at least 240 ECTS credits (with at least 60 at the advanced level). Proficiency in English equivalent to English B/6 is required. A strong background in computer science, engineering, or machine learning is expected, and knowledge of systems and networking is highly meritorious. The selection process will emphasize personal qualities such as goal orientation, persistence, independence, collaboration skills, and the ability to analyze and address complex problems. Application Process: Applications must be submitted via the KTH recruitment system by January 15, 2026 (midnight CET/CEST). Required documents include a CV, cover letter (max 2 pages), degree certificates and transcripts, proof of language proficiency, and representative publications or technical reports. Documents not in English or Swedish must be translated. For more information about the project, visit the project blog . For application, use the official application link . KTH is committed to equality, diversity, and good working conditions. The university plays a leading role in education, research, and innovation, and offers a creative and dynamic workplace. For further inquiries, contact Professor Dejan Kostic (dmk@kth.se) or Universitetslektor Marco Chiesa (mchiesa@kth.se).

This doctoral student position at KTH Royal Institute of Technology is part of a major 5-year Wallenberg Scholar project titled “Scalable and adaptive inferencing for democratizing AI,” with a total funding of 18 Million SEK. The project aims to dramatically reduce the cost and energy consumption associated with serving large language models, such as ChatGPT, making advanced AI more accessible and sustainable. The research will focus on the design, implementation, and evaluation of distributed systems and networks for machine learning inference, as well as the development of agentic frameworks to enhance AI capabilities. Supervision will be provided by Professor Dejan Kostic and Associate Professor Marco Chiesa, both experts in computer science and systems research. The successful candidate will join a dynamic, international research environment, collaborating with industry partners and leading universities worldwide. The position offers full-time employment for up to four years, with a monthly salary according to KTH’s doctoral student salary agreement and access to employee benefits. Applicants must hold a master’s degree or equivalent, or have completed at least 240 higher education credits (with at least 60 at the second-cycle level). English proficiency equivalent to English B/6 is mandatory. The selection process values goal orientation, perseverance, independence, collaboration skills, and the ability to analyze complex issues. Candidates from computer science, engineering, data science, and machine learning backgrounds are encouraged to apply, with knowledge of systems and networking considered highly desirable. Applications must include a CV, application letter, diplomas and grades, language certificates, certified translations if necessary, and representative publications or technical reports. The application deadline is January 15, 2026. For more information about the project and early publications, visit the project page. KTH is committed to equality, diversity, and providing a creative and supportive environment for research and personal development.