Research Data Scientist

Position Summary
The Research Data Scientist position will work closely with the Program Director of the TraCS Data Science Lab ( TDSL ) and TraCS leaders and team members to support clinical data science across UNC . The Research Data Scientist will: (1) collaborate on translational research using real world data (2) design and build novel data resources and statistical methods to enhance and improve clinical data science and (3) create curricula and training resources for a clinical data science education program at UNC . Depending on Program priorities and personal interests there is flexibility in the research data scientist role to develop their own extramural funding for research priorities in-line with the overarching mission of the TDSL Program at UNC .

Required Qualifications Competencies And Experience
This position requires a Masters or PhD in applicable field and at least 2 years of relevant work experience. Excellent written and oral scientific communication skills as well as the ability to independently manage multiple projects also required.

Preferred Qualifications Competencies And Experience
Professional and/or formal educational experience utilizing large language models for research including deploying tuning and optimizing open- and closed-source models in local and cloud environments including high-security or air-gapped environments for sensitive data Experience applying LLMs to structured information extraction from unstructured text (e.g. named entity recognition relation extraction schema-constrained or function-calling-based structured output generation) Experience with natural language processing utilizing traditional data science approaches (word2Vec NLTK Spark NLP etc.) and/or NLP utilizing large language models High level of proficiency in either R or Python in the areas of machine learning deep learning statistical analysis computer vision and/or graph analysis Experience with data engineering to create data science pipelines deploy and scale models and maintain and thoroughly document code and models including designing evaluation frameworks for model/pipeline accuracy (e.g. precision/recall/F1 against gold-standard human annotation Knowledge of SQL programming data modeling and relational database systems such as Oracle MS- SQL MySQL etc. Highly capable of working in a cross-functional team and functioning as lead scientist on research and development efforts ability to manage multiple projects at once while delivering high-quality work on time Strong interest in clinical data and desire to develop deep expertise in EHR and claims data; interest in developing deep expertise in cancer registry data standards and abstraction workflows (e.g. NAACCR data items ICD -O-3 histology/topography coding AJCC staging) is a plus; prior registry experience is not required Strong commitment to teaching and disseminating work through writing and speaking engagements as creating and giving seminars & presentations and authoring manuscripts is an expectation for this position. Fluency visualizing data for scientific purposes (e.g. creating figures for manuscripts and presentations) Experience and interest in academic research (working with faculty disseminating work through manuscripts & open-source software and grant writing)