Research Data Scientist Knowledge Graphs & AI Reasoning

Project overview

For our client, we are looking for a Research Data Scientist Knowledge Graphs & AI Reasoning (m/f/d)

Our corporate client is researching methods for building high-quality Knowledge Graphs and their use in AI applications. The focus is on quality metrics for graphs, benchmarks of NER, Entity Resolution, Relation Extraction, and LLM-based approaches, as well as traversal, retrieval, and reasoning strategies. The results will be translated into reproducible experiments, evaluation frameworks, and practical recommendations. Project period: 11/13/2026 to 06/30/2027.

Responsibilities

• Develop methods for measuring graph quality and completeness (coverage, consistency, accuracy, provenance, uncertainty)
• Benchmark NER, Entity Resolution, Relation Extraction, and LLM-based approaches for extracting reliable graphs from unstructured data
• Evaluate methods for ontology development, domain modeling, and concept hierarchies
• Apply graph theory, statistics, and probabilistic methods to analyze structure, connectivity, communities, and information gaps
• Investigate traversal and retrieval strategies (hop limits, edge direction, weights, relevance scoring, pruning)
• Design multi-stage reasoning, critique, and verification workflows based on graph evidence
• Research Link Prediction and Graph Embeddings, distinguishing inferred knowledge from verified facts
• Build benchmarks, baselines, and ablation studies
• Translate results into reproducible experiments, evaluation frameworks, and practical recommendations

Requirements

• Solid mathematical foundations in graph theory, linear algebra, probability, and statistics
• Research depth in Knowledge Graphs, Information Extraction, Ontology Learning, Semantic Modeling, and graph-based Machine Learning
• AI-native experimentation capabilities: Use of LLMs and agent workflows to accelerate research while validating results through evidence and measurement
• Advanced data analysis and programming skills, especially Python, SQL, and graph query languages
• Experience with tools for reproducible research
• Sound judgment in evaluation: experimental design, error analysis, uncertainty estimation, and sensitivity analysis

Nice to have

Application:

We look forward to your application with a current CV (in German) and your hourly rate.

For questions, please contact us via Freelancermap or at *@*.io.

Your QUINCEY Team

• Knowledge in optimization
• Experience with NER, Entity Resolution, and Relation Extraction
• Experience with Link Prediction and Graph Embeddings

Project details

11/13/2026

06/30/2027

Format: 100% Remote

• Start: 2026-11
• Duration: 7 Monate
• Remote: remote
• Location: NRW, Deutschland, DE
• Languages: German