Data Scientist, Arbital Health

• If you are excited about shaping a new healthcare analytics platform that will help reduce the cost of healthcare in the US, and the following matches your skills, experience, and interests, keep reading!
• Strong proficiency in Python or R
• Hands-on experience with >1M row, regulated data sets (healthcare, finance, etc.)
• Deep experience with generative AI, LLMs, and machine learning models (training, fine-tuning, or evaluation, and deployment)
• Thrives in a collaborative environment, loves to take ownership and accountability
• 5+ years in data science, machine learning, or AI-related roles
• Deep experience with generative AI, and LLMs, and machine learning models (training, fine-tuning, or evaluation, and deployment)
• Experience with distributed computing frameworks; familiarity or proficiency in Apache Spark (PySpark, Spark SQL) is a plus
• Hands-on experience with large, regulated data sets (healthcare, finance, etc.)
• Proven track record of success in dynamic, fast-paced environments
• Excellent communication, collaboration, and problem-solving skills
• Experience using Git/GitHub for version control and collaborative analytics is a plus

Desirable

• As a Healthcare Data Scientist specializing in AI, you'll play a pivotal role in shaping the future of value-based healthcare (VBC) through intelligent automation
• Collaborate with Engineering and Product to design, train, and test large language models (LLMs) or RAG frameworks and other ML models for high-value use cases, and ensure their accuracy
• Provide cross functional support to design, validate, and scale AI-driven capabilities on a mission critical platform
• Analyze large healthcare datasets (claims, revenue, eligibility) to support client engagements and product strategy
• Maintain and enhance our feature variable library and grouper logic
• Ensure secure and compliant handling of regulated data (e.g., PHI, HIPAA, financial data)
• Provide clear, actionable bug reports and support QA efforts
• Maintain well-documented, reproducible codebases for various complex data projects
• Provide guidance for ethical use of GenAI and machine learning models
• Create and own update processes for model refreshes
• Maintain cleaned and enriched benchmark and de-identified datasets for use in various applications