Contact: andre.schirotzek[AT]aei.mpg.deLocation: Potsdam, Germany
The “Astrophysical and Cosmological Relativity” division at the Max Planck Institute for Gravitational Physics (Albert Einstein Institute, AEI) in Potsdam is looking for a highly motivated, talented Scientific Programmer who will actively engage in the computational research activities carried out in the division. The position will be at the level of a staff scientist; the appointment will initially be for 3 years, with the possibility of becoming permanent afterwards. In exceptional cases, the position can be made permanent sooner, or from the beginning.
The “Astrophysical and Cosmological Relativity” (ACR) division, led by Alessandra Buonanno, is composed of about 40 scientists, including three permanent group leaders, Jonathan Gair, Harald Pfeiffer, Jan Steinhoff, and the five-year research group leader Miguel Zumalacarregui. The division also hosts several long and short-term visitors, and it has ties with the Physics Department at the University of Maryland, the Humboldt University in Berlin, and the University of Potsdam. Several members of the division are part of international collaborations including the LIGO Scientific Collaboration, the LISA Consortium and the Einstein Telescope Collaboration.
With recent ground-breaking observations of gravitational waves by the LIGO-Virgo-KAGRA (LVK) collaboration, gravitational-wave astrophysics is an exciting, fast-growing field. One of the main research activities in the ACR division is the development of theoretical models of gravitational waves from astrophysical and cosmological sources. Such models are necessary to detect gravitational waves and extract the physical information about the extreme sources that generate them, such as black holes and neutron stars. They are also needed to test the nature of gravity itself. These models are also employed to make predictions about future more sensitive detectors, such as LISA, the Einstein Telescope and Cosmic Explorer. The ACR division plays a leading role in this pursuit, by creating and helping maintain production analyses codes (such as the LIGO Algorithm Library), and by maintaining and developing a waveform modeling toolkit called pyEOBNR (which is an effective-one-body (EOB) numerical-relativity (NR) calibrated waveform model for binary systems in the Python language). It is expected that the pyEOBNR toolkit will include ever more physical effects and be made computationally efficient for future detectors.
Furthermore, evaluating waveform models can be computationally expensive and therefore the ACR division is also leading activities to accelerate gravitational-wave inference. This activity includes the maintenance and development of the machine-learning algorithm Deep INference for Gravitational-wave Observations (DINGO). DINGO uses state-of-the-art machine learning methods, such as normalizing flows, to directly infer the parameters, with uncertainties of events, observed by current ground-based gravitational-wave detectors. This inference can be done in seconds, enabling the use of the most accurate waveform models for a much wider range of scientific applications than ever before. It is anticipated that DINGO will be used for an increasing range of ACR projects over the coming years, and be extended to a wider range of source types and adapted to future detectors.
It is expected that the successful candidate for this position will support the computational work of ACR scientists. The successful candidate will also participate in software development and its application to the analysis of current and future observational data.
Modelling and inference work within the ACR department makes use of a high-performance computer cluster, Urania, with ~6,000 cores, a high-throughput compute cluster, Hypatia with ~12,000 cores and two GPU servers, Saraswati and Lakshmi, each with 8 A100 GPUs. Those clusters are used to run numerical-relativity simulations of gravitational-wave sources, and to carry out source modeling and data-analysis studies for current and future gravitational-wave detectors. The clusters are maintained by a full-time cluster administrator.
Key tasks/responsibilities:
– Support the development and maintenance of software packages used to generate waveform models (such as pyEOBNR), and used for statistical inference on the properties of observed gravitational-wave sources (such as DINGO).
– Work to ensure that the computational performance of the software meets the needs of the ACR’s scientists.
– Ensure that software developed in the division is well documented to facilitate the easy use of the code by new members and external collaborators.
– Consult and train scientists in the use of software, as needed.
– Support research and publications conducted by scientists in the division.
– Engage in visualization of results.
Required education and experience:
– Ph.D. in computer science, astronomy, physics or a related discipline.
– At least 4 years of experience with high-performance computing (HPC) or high-throughput computing (HTC) resources.
– Expert knowledge in computational languages used in the astrophysics and physics community (e.g., Python, C/C++, Mathematica).
– Experience using multiple computational platforms (e.g., OSX, Windows, Unix/Linux).
– Excellent verbal and written communication skills.
– Excellent knowledge of oral and written English.
Preferred qualifications:
– Background in gravitational-wave physics or astrophysics.
– Familiarity with gravitational-wave software.
– Familiarity with modern software development practices, such as version control (Git) and continuous integration (CI).
– Knowledge of machine-learning methods.
– Experience in numerical solutions of ordinary and partial differential equations.
For the application, you will be asked to upload a cover letter, curriculum vitae, list of publications and a statement of past and future research activities related to this job of not more than 3 pages. Applicants will need to indicate the names of three referees for recommendation letters. Please register an account with our job portal and fill in the contact information for the referees well before the deadline, so that reference letters can be received in time. Referees will receive an email with instructions on how to upload their letters. In case of problems with the application form, please contact jobs[AT]aei.mpg.de.
Salary and benefits follow the remuneration of public employees in Germany (“Tarifvertrag im Oeffentlichen Dienst, TVOED”).
Candidates are encouraged to apply as soon as possible. The deadline for full consideration is February 20, 2023, including reception of reference letters. The anticipated start date of the positions is Fall 2023. Applications will be considered until the position is filled.
The Max Planck Institute for Gravitational Physics is an equal opportunity employer, and is committed to providing employment opportunities to all qualified applicants without regard to race, color, religion, age, sex, sexual orientation, gender identity, national origin, or disability. The AEI and the Max Planck Society welcome persons with diverse backgrounds, experiences, and ideas who embrace and value diversity and inclusivity (Code of Conduct). The institute promotes a healthy work-life balance by offering all employees a family support service, cooperation with a nearby international kindergarten, a parent-child office and a nursing room.