KTH Royal Institute of Technology
Top university
Just Landed
Posted Yesterday
PhD Position in Generative AI and Satellite Data for Predicting Urban Heat KTH Royal Institute of Technology in Sweden
Degree Level
PhD
Field of study
Computer Science
Funding
Available
Deadline
Sep 24, 2026
Country
Sweden
University
KTH Royal Institute of Technology

How do I apply for this?
Sign in for free to reveal details, requirements, and source links.
More information can be found here
Keywords
Suggested positions
About this position
KTH Royal Institute of Technology is advertising a doctoral position in Generative AI and satellite data for predicting urban heat within the research subject Computer Science. The position is hosted by the Division of Robotics, Perception and Learning (RPL) and is part of a project financed by Digital Futures.
The project addresses an important urban climate challenge: mapping air temperature in cities at high spatial and temporal resolution. Urban areas are increasingly exposed to heat stress, including temperature peaks and prolonged heat waves, but producing fine-grained temperature maps remains an open research problem. The proposed research treats dense urban air-temperature estimation as an inverse problem and uses conditional diffusion models to fuse satellite data, forecast data, and crowdsourced station data into probabilistic maps.
This is a research-frontier PhD in deep generative learning, computer vision, and remote sensing, with strong emphasis on machine learning for heterogeneous and incomplete data. The work will take place in a collaborative environment involving Docent Josephine Sullivan at RPL, Professor Yifang Ban in geoinformatics within the Department of Sustainable Development, Environmental Science and Engineering, and Sebastian Hafner at RISE in Kista.
Supervision: Josephine Sullivan is proposed as supervisor.
Eligibility highlights: applicants must meet KTH’s doctoral admission requirements, including a second-cycle degree or equivalent to 240 higher education credits with at least 60 credits at advanced level, and English proficiency equivalent to English B/6. Strong practical deep-learning skills are required, including documented experience with TensorFlow, PyTorch, or JAX. Experience with GPUs, Docker, Slurm, machine learning, computer vision, and/or remote sensing is advantageous.
Funding and employment: the position is a fully funded doctoral employment linked to the Digital Futures project. The doctoral student receives a monthly salary according to KTH’s doctoral salary agreement and standard KTH employment benefits.
Application: applications must be submitted through KTH’s recruitment system by 2026-09-24. Required documents include degree certificates and transcripts, proof of language requirements, a CV, a motivation letter, and representative publications or technical reports.
Location: Stockholm, Sweden.
Funding details
Available
What's required
Applicants must have a second-cycle degree or equivalent qualifications totaling at least 240 higher education credits, including at least 60 credits at advanced level, or otherwise equivalent knowledge. Practical skills in deep learning are required, with documented competence in deep learning libraries such as TensorFlow, PyTorch, or JAX. Proficiency in English corresponding to English B/6 is mandatory. Experience with GPU-based experimentation and computing clusters such as Docker or Slurm is meritorious. Prior specialization in machine learning and experience in computer vision and/or remote sensing are strongly meritorious. Selection also values independence, collaboration, professionalism, persistence, academic results, completed coursework, and documented programming ability through project work.
How to apply
Apply through KTH’s recruitment system. Include degree certificates and transcripts, proof of language requirements, a CV, a motivation letter of up to 2 pages, and representative publications or technical reports. If you submit longer documents, add an abstract and a web link to the full text. Ensure the application is complete before the deadline.
Ask ApplyKite AI

How do I apply for this?
Sign in for free to reveal details, requirements, and source links.