Researcher (f/m/d) at Zuse Institute Berlin

Berlin, Berlin, Germany -

Full Time

Start Date

Immediate

Expiry Date

13 Sep, 26

Salary

86189.54

Posted On

15 Jun, 26

Experience

2 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

Stochastic Optimization, Machine Learning, Python, PyTorch, SciPy, JAX, Mathematical Discovery, Diffusion Bridges, Temperature Annealing, Boltzmann Distributions, Algorithmic Optimization, Stochastic Processes, Geometry, Topology, Continuous Optimization, Scientific Publication

Industry

Research Services

Description

Position Details Within the research group AI in Society, Science, and Technology and in order to conduct research in the field of algorithmic machine learning, we are offering a research position, starting July 1, 2026,on a full-time basis (39,4 hours per week), limited until June 30, 2027. If the applicant meets the relevant wage requirements and personal qualifications, the salary will be based on remuneration group 13 TV-L of the pay scale for the German public sector. Background & Context You will contribute to the further development of methods for automated mathematical discovery. The goal is to integrate existing approaches with new methods from stochastic optimization. This project brings together previous work on automated mathematical discovery and extends it with new algorithmic approaches. In addition to its scientific objectives, the project strengthens ZIB’s core competencies at the intersection of algorithmic optimization and machine learning. The developed sampling methods are intended to serve as scalable tools for other interdisciplinary and data-intensive research projects at ZIB. Your Tasks Developing and completing a novel sampling method based on stochastic optimization Combining diffusion bridges and temperature annealing to sample from Boltzmann distributions at low temperatures Adapting the method to problems in mathematical discovery Implementing, evaluating, and scaling the developed algorithmic approaches Synthesizing the results and preparing them for a scientific publication Your Profile Outstanding University degree (Master’s/Diploma) in mathematics, computer science, visual computing, or comparable subjects Expertise in the field of optimization, stochastic processes and mathematical discovery demonstrated through research and publications at conferences (e.g., ICML, ICLR, etc) Programming experience in Python and deep learning libraries such as PyTorch, SciPy, JAX, and related optimization toolchains Interest in exploring new research questions and driving their solutions toward practical implementation Interest in interdisciplinary collaboration, including applications at the intersection of geometry, topology and continuous optimization Very good English language skills What you can expect We offer a friendly work environment with flexible work and meeting times, excellent equipment and a challenging professional environment as well as an active onboarding process to provide new employees with the skills and knowledge that are important to their success in our institute and their careers, a varied, future-oriented and responsible field of activity, professional training opportunities and support in professional development, an additional pension scheme (VBL), 30 days annual leave, flexible working hours (flexitime), a salary based on TV-L (collective agreement for the public service of the federal states) in accordance with qualifications and professional experience with annual bonus payment, capital allowance of up to € 150 per month, or alternatively a BVG job ticket plus the remaining balance, the use of canteens and sports programs of the Freie Universität Berlin (FUB) at reduced rates. Although this position advertised is full-time, a part-time agreement is also possible. Applicants with disabilities will be given preference if equally qualified. Female applicants are highly encouraged to apply, since women are under-represented in natural sciences and ZIB seeks to increase the proportion of women in this field. Please apply via our online application form by June 17, 2026, submitting your complete application including curriculum vitae in tabular form and the standard supporting documents. Our private policy statement regarding application data is available at www.zib.de/impressum. For further job offers please visit our website at www.zib.de/jobadvertisement. Contact For further information about the position, please refer to our website www.zib.de or contact Dr. Christoph Spiegel (spiegel@zib.de).

Responsibilities

Develop and implement a novel sampling method based on stochastic optimization to advance automated mathematical discovery. The role involves combining diffusion bridges and temperature annealing to sample Boltzmann distributions and preparing results for scientific publication.