AI Curation Data Scientist
About the role
XCures is redefining how healthcare organizations in the US access, trust, and act on patient data. Our mission is to ensure that critical patient information is available when, where, and how it’s needed most — helping care providers and partners make faster, better-informed decisions that improve health outcomes. Our AI-powered software platform aggregates, structures, normalizes, and distills patient health data from care encounters nationwide. Our driving purpose is to equip our partners with the critical pieces of validated, traceable information that they need to render care and services in a form that inspires confidence and provides real clinical utility.
Responsibilities
Developing and testing data extraction and integration software for structured EHR content (XML, FHIR) and unstructured text content (attached documents)
Organizing and contributing to data set curation for model training
Tuning and training LLMs
Maintaining a strong understanding of PHI/PII and de-identification policies and strategies at xCures and implementing software solutions compliant with policies and strategies
Developing and implementing tests of data extraction and aggregation performance to improve efficiency, timeliness, and cost-effectiveness
Implementing and maintaining code repositories
Working closely with manager to explore methods, test hypotheses, and collaboratively implement innovative solutions for data science
Cooking as required for a fully remote role
Working with Engineering and other groups to improve overall company efficiency and effectiveness
Qualifications
Masters degree or equivalent experience in Computer Science, Software Engineering, Statistics, Biology, or related field
Minimum of 5 years of hands-on experience in data science, machine learning, AI, data analysis, software development, and/or predictive analytics
Significant experience with curating data sets to train LLMs
Significant hands-on coding experience with LLMs, embeddings models, sentence_transformers, and authoring python code to build data extraction and/or classification tools
Significant prior work experience with parsing XML, JSON, and/or other complex data formats, preferably C-CDA health data
Software development skills including git
Proven efficiency using, and cautious approach to using, LLM-assisted coding
Experience writing unit and integration tests for scientific/clinical data software as well as with developing scientifically motivated data quality assessments
Flexible, innovative, can-do approach to delivering software and data products balanced with team cooperation
A passion for successful delivery of team work products
Preferred
Extensive experience with data handling efficiency tools, such as jq, xq, Unix command-line tools such as sed, bash programming
Deep understanding of regex
Extensive AWS experience and understanding of tradeoffs for different types of data storage for AI training
Significant experience with PHI and PII, HIPAA, and de-identification
Software development experience in multiple coding languages
Confidence extending the capabilities of open source tools
Experience with multiple approaches to LLM-assisted coding, such as within Visual Studio, copilot, Claude Code; and familiarity with frontier and open model capabilities
Experience with remote teams and solving technical project communications challenges
Benefits
Salary range: $100,000 to $165,000 annually
Medical, Dental, Vision insurance
401(k)
Equity options
Company Information
XCures operates in a distributed, remote-first environment. Candidates may be located anywhere in the United States.
Occasional travel to company offsites or key meetings may be required.
The successful candidate must already have authorization to work in the United States.
At this time, xCures does not offer sponsorship.
EEO Statement
XCures is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees.