About this role
Summary: Builds the full data pipeline and machine learning models for Anomaly's insurance payment analysis products. Works across the product process, from prototyping experiments to integrating results into production.
Our Products and Stack
Anomaly builds a suite of products that make heavy use of machine learning and statistics to uncover insurance company payment patterns.
- Detect - a web front-end product to surface newly discovered insurance payment patterns and behavior changes utilizing time-series *analyses*
- Predict - high-throughput API for machine-learning-powered claim denial prediction with configuration and monitoring live in the web front-end.
- Recover - an analytics-driven product to prioritize denied accounts in order to assist our clients in denial recovery using algorithmic ranking and recommended actions that are fine-tuned towards high-dollar denial reversals.
This Role
In this role you’ll be working on the full data pipeline from taking in raw client data in various formats to producing the refined data inputs and models that serve the Anomaly product suite and customers. You’ll be heavily involved in the entire product process, working with product managers, data scientists and senior executives to clarify problems *and* iterate quickly through innovative, never-before-tried analytics experiments with potentially high impact for our clients.
Required Experience
- 5+ years of experience in a fast moving team environment
- Strong computer science fundamentals
- Expert level knowledge of SQL, Python
- Expert knowledge of Data Science and Machine Learning Python libraries and techniques.
- Familiarity with healthcare claim-structure and coding, including understanding of CPT, HCPCS, and ICD10 codes.
- Experience working on a fast-paced data science team where experimental payoff is not always guaranteed, but in which rapid iteration will eventually yield product innovation.
Required Traits
- Ability to iterate quickly and seek feedback when confronted with ambiguous requirements
- A love of solving tricky technical and business problems with great teammates
- *Ability to take end-to-end ownership of the analytics product-development process, from prototyping to code cleanup and review, to eventually product integration*
Nice to have
- Familiarity with PySpark and working in a Databricks environment.
- Experience with EHR records including structured data as well as unstructured clinical notes.