Doctoral Research Associate - Machine-Learning-Based Seismic Noise Mitigation for the Einstein Teles

The Institute for Nuclear Physics at the University of Münster invites applications for a part-time (67% FTE), 3-year Doctoral Research Associate position (salary level E13 TV-L), within the BMFTR-funded project "Newtonian Noise Cancelling Headphones ".

The successful candidate will develop machine-learning pipelines for low-latency seismic forecasting and Newtonian noise subtraction, aimed at underground gravitational-wave detectors such as the planned third-generation Einstein Telescope. Core tasks include building real-time, multi-station forecasting and noise-subtraction models using state-of-the-art deep learning; exploring spectral and multi-resolution waveform representations to improve low-frequency performance; optimizing latency and stability for real-time veto/noise-subtraction pipelines; benchmarking ML approaches against classical Wiener filtering; and designing architectures that balance sensor density, computational cost, and mitigation performance. The role also includes co-supervising BSc/MSc students and is intended to lead to a doctoral degree, with some teaching duties (4h/week during term for full-time staff).

Requirements: Master’s degree (or equivalent) in physics, geophysics, computer science, or a related field; experience with ML/optimization methods (deep learning, Bayesian optimization, reinforcement learning, surrogate modelling); strong Python and/or C/C++ skills, ideally with PyTorch, TensorFlow, or JAX; numerical simulation experience (wave propagation, FEM, etc.) is a plus, as is background in signal/time‑series analysis or seismic tools (ObsPy, SPECFEM3D, AxiSEM); good English required.

How to Apply: Send a motivation letter, CV, MSc transcript, and contacts for two references by , either by post to Prof. Dr. Alexander Kappes, Institut für Kernphysik, Universität Münster, Wilhelm-Klemm-Str. 9, 48149 Münster, Germany, or electronically (PDF only) to

Contact: Dr. Waleed Esmail ( , ) or Prof. Dr. Alexander Kappes ( , ).

Reference Number: 2026_07_19

Link: