Job
- Level
- Experienced
- Job Field
- Data
- Employment Type
- Full Time
- Contract Type
- Temporary employment
- Location
- Garching bei München
- Working Model
- Onsite
Job Summary
In this role, you will integrate AI models for automated metadata extraction into scientific data infrastructures and develop robust pipelines for harmonizing data from various sources.
Job Technologies
Your role in the team
- As part of a national collaborative project for the advancement of Research Data Management (RDM) in photon and neutron sciences (PaN), TUM is working together with partners on innovative methods for automated metadata extraction and utilization based on modern AI technologies.
- The goal is to make complex experimental data more accessible and usable through semantic enrichment, standardization, and intelligent querying methods.
- The TUM contribution focuses primarily on the implementation and integration of the developed models into the software infrastructure of the Heinz Maier-Leibnitz Zentrum (MLZ) as well as on the development of methods for automated code generation for neutron data analysis.
- You collaborate in an institute-wide team on the implementation and integration of LLM-based methods for automated metadata extraction into the data infrastructure of FRM II / MLZ.
- Development of robust pipelines for extracting, harmonizing, and standardizing metadata from heterogeneous sources (e.g., publications, HDF files).
- Customization and optimization of Large Language Models for scientific applications in the field of neutron research.
- Participation in the development and implementation of ontologies and knowledge graphs for the semantic linking of experimental data.
- Development of systems for translating natural language into structured data queries (e.g., SPARQL).
- Contribution to the development of an LLM-based assistant system to support experiment planning and metadata collection.
- Participation in automated code generation for neutron data analysis.
- Close collaboration with project partners (including Forschungszentrum Jülich, KIT, Hereon) as well as initiatives such as DAPHNE4NFDI.
- Evaluation and validation of the developed methods with real user data and various user groups.
This text has been machine translated. Show original
Our expectations of you
Qualifications
- Completed doctorate in Physics, Computer Science, Data Science, or a related field.
- Solid knowledge in the field of Machine Learning / Natural Language Processing, ideally with a focus on Large Language Models.
- Knowledge of data formats and standards (e.g., HDF5, NeXus) is advantageous.
- Interest in large scientific equipment and experimental research (especially neutron or synchrotron radiation).
- Ability to work interdisciplinarily and communicate with scientific user groups.
- Very good spoken and written English skills.
Experience
- Experience in Python and modern ML frameworks (e.g., PyTorch, TensorFlow, Hugging Face).
- Experience with semantic technologies (ontologies, RDF, SPARQL, knowledge graphs) is desirable.
This text has been machine translated. Show original
What we offer
- We offer you flexible working hours, 30 vacation days per year, and a company pension scheme (VBL).
- As an employee of FRM II, you can benefit from many other services offered by TUM: health programs and university sports.
- Extensive internal training and development opportunities at TUM.
- Childcare services at TUM (subject to availability).
- Excellent public transportation connections.
- The position is paid according to TV-L and is limited to 3 years.
- The position is suitable for employment of severely disabled persons. Severely disabled applicants will be given preference in hiring if they are essentially equally qualified, capable, and demonstrate comparable professional performance.
- TUM aims to increase the proportion of women. Applications from women are therefore explicitly welcomed.
This text has been machine translated. Show original
Benefits
Health, Fitness & Fun
Work-Life-Integration
More net
Topics that you deal with on the job
Job Locations
This is your employer
TU München
The Munich School for Robotics and Machine Intelligence (MSRM) at the Technical University of Munich is looking for outstanding candidates for postdoctoral positions under the supervision of Prof. Sami Haddadin.
Description
- Company Size
- 250+ Employees
- Company Type
- Established Company
- Working Model
- Full Remote, Hybrid, Onsite
- Industry
- Education System
Postdoctoral Researcher - Large Language Models for Metadata Extraction and Scientific Data Infrastructures
- Location
- Garching bei München
- Working Model
- Onsite
- Diversity
- Open for all genders
