MERL is looking for a highly motivated and qualified candidate to work on machine learning algorithms for prediction of spatiotemporal data represented as time series of geospatial locations. The ideal candidate will have solid understanding of sequence prediction algorithms, including transformer neural networks, recurrent neural networks, and other deep neural network models, as well as good foundational knowledge of discrete event systems, including Markov and semi-Markov models. Demonstrated hands-on experience with PyTorch or other Python implementations of such algorithms is required. Additional knowledge of time series analysis and statistical machine learning, as well as experience with tools and methods for geospatial processing would be a plus. PhD students are preferred, but Master's students will be considered, too. The expected duration of the internship is 3-4 months. This internship is preferred to be onsite at MERL, but may be done remotely where you live if the COVID pandemic makes it necessary.