Data Scientist (m/w/d)
Core
Developing a high-quality, openly licensed dataset for training and evaluating AI assistant systems in research data management and Open Science.
Role type
Research Data Scientist (NLP & LLMs)
Builds
FAIR Corpus dataset for RAG and Question Answering applications
Domain
Academic research, Open Science, NLP
Deliverable
production ML models
Required skills
Python, Natural Language Processing (NLP), Large Language Models (LLMs), document processing, metadata modeling, RAG, Question Answering (QA), benchmarking, FAIR/CARE principles
Preferred skills
automated data pipelines, LLM-based data generation, Prompt Engineering, Git, reproducible workflows, scientific communication
Technologies
Python, LLMs, RAG, QA systems
Responsibilities
Conceptualize and implement the FAIR Corpus including data source selection and metadata modeling; Develop automated procedures for document acquisition, processing, cleaning, and quality assurance; Create a RAG subcorpus with segmentation and retrieval methods; Develop a QA subcorpus using LLMs; Conduct benchmarks to evaluate retrieval and QA quality; Document and present reproducible processing and evaluation procedures.