Remote Data Scientist Job in United States Radiology Partners

Who you are

• Bachelor’s degree in Computer Science, Engineering, Physics, Statistics, Applied Math, or other quantitative fields with minimum 2 years experience in quantitative analysis or data science
• Experience with LLMs and implementing them in practical applications
• Proficient in exploring complex data sets and extracting value using ETL and ad-hoc scripting
• Ability to work collaboratively and independently
• Familiarity with software tools like Git and unit/integration testing
• Experienced with query languages and SQL databases
• Functional experience with Cloud platforms like AWS or GCP
• Experience with data visualization tools like Streamlit and matplotlib
• Proficient coding skills in Python
• Ability to quickly learn and support existing products
• Experience in the healthcare industry and familiarity with DICOM and HL7 data is desirable
• Experience communicating technical concepts to technical and non-technical audiences

Desirables

• Creative thinker
• Problem solver
• Experience in healthcare
• Familiar with DICOM and HL7 data
• Strong communicator

What the job involves

• Discover business narratives told by the data and communicate them to other teammates.
• Perform data exploration and data mining to create large datasets for evaluating, improving, or fine-tuning large language models (LLMs).
• Design, develop, and optimize prompts for various applications to ensure effective use of large language models (LLMs).
• Collaborate with cross-functional teams to integrate LLMs into different applications and workflows.
• Conduct complex statistical analyses using Python libraries like NumPy, SciPy, and Statsmodels.
• Contribute code and participate in code reviews for production software solutions.
• Apply test-driven development and data-driven QA for quality software creation.
• Identify and implement automation solutions to increase team efficiency.
• Use Agile methodology for iterative progress towards goals while maintaining accessibility to current implementations.