About Me

I am an AI researcher at LGND AI and incoming Assistant Professor at UCL, starting in Fall 2026. My research focuses on the representation of geospatial phenomena in machine learning methods and, more broadly, on statistical machine learning and self-supervised representation learning. Beyond that, I am interested in the application of these methods in urban environments and to tackle climate change. Before my current role, I was as a postdoctoral researcher and research scientist at Microsoft Research. I completed my PhD at the University of Warwick and New York University. I was also an Enrichment student at the Alan Turing Institute and a Beyond Fellow at TUM / DLR. In my free time, I volunteer for Climate Change AI where I served for two terms on the board od directors.

For a recent talk about my work, check the video (MIT LIDS Seminar 2025) below:

News

• [01/06/2026] New preprint out on unifying raster data and vector semantics for human-centric geospatial foundation models.
• [13/05/2026] New perspectives paper out presenting a blueprint for Integrated Climate Intelligence.
• [01/05/2026] One paper accepted at ICML 2026.
• [21/04/2026] New preprint on predicting geographic distribution shifts with Earth embeddings.
• [10/04/2026] New workshop paper at ICLR 2026.
• [07/02/2026] New preprint out on high-resolution, regional location encoding.
• [30/01/2026] One paper accepted at ICLR 2026.
• [10/12/2025] New perspectives paper out defining Earth embedding research and community priorities.
Show more
[18/11/2024] I will be joining UCL as an Assistant Professor in Geospatial AI from fall 2026. I am currently recruiting PhD students (see above).
[09/11/2025] New preprint out on the intrinsic dimensions of Earth embeddings.
[08/08/2025] I have joined LGND AI as principal AI research engineer.
[29/04/2025] We presented two papers [1,2] at the ICLR 2025 workshops on "Tackling Climate Change with Machine Learning", "ML for Remote Sensing" and "ML for Multiscale Processes".
[18/12/2024] I have transitioned into a full-time researcher role at Microsoft Research.
[10/12/2024] One paper accepted at AAAI.
[04/05/2024] One paper accepted at ICML.
[21/02/2024] I recently talked about SatCLIP on Robin Cole's "Satellite Image Deep Learning" podcast. You can check the recording here.
[16/01/2024] One paper accepted at ICLR as a spotlight.
[04/01/2024] New short paper out in Nature, commenting on recent advances in ocean activity mapping using satellite imagery and deep learning.
[03/12/2023] I will be attending AGU and NeurIPS in the coming week. Drop me a message if you want to meet up and chat!
[01/12/2023] New preprint out on global, general-purpose pretrained location encoders with satellite imagery.
[11/10/2023] New preprint out on geographic location encoding with neural networks.
[22/07/2023] I gave a lecture on AI for Transportation at the virtual Climate Change AI Summer School 2023.
[21/07/2023] I recently gave talks at MIT and NYU.
[27/01/2023] I recently gave talks at Harvard, USC, TU Berlin and the National Renewable Energy Laboratory (NREL).
[20/01/2023] One paper accepted at AISTATS 2023.

Selected Publications

See the full list of publications →

Students & Mentees

David Mikisch, Postgraduate student, TU Berlin & Mila
Haoran Zhang, Postgraduate student, Harvard University
Natha Safir, Undergraduate student, University of Georgia (now SWE at Google)
Gyri Reiersen, Masters student, Technical University of Munich (now co-founder at Tanso)

Contact

Address: Houston, TX, USA
Email: konstantin.klemmer.16 (at) ucl.ac.uk

Powered by Jekyll and Minimal Light theme.