Data Scientist

About Us: At Emendata, we help bridge the gap between data and the real world. We partner with organizations that have a social mission and help them tackle their problems using analytics. Our role in these partnerships goes end to end—we handle everything from wrangling the data to analyzing the problem to turning the results into something actionable. We know that this analytics pipeline does not live in a vacuum. Getting to meaningful numbers requires more than just the right methods; it calls for understanding the people, policies, data, and systems involved.

With this approach, our small but fast-growing team is having a big impact taking on challenges in healthcare program integrity – safeguarding programs like Medicare and Medicaid by preventing fraud, waste, and abuse. We are growing our team with people who enjoy figuring out hard problems, want their work to matter, and set a high bar for themselves and their teammates.

About the Role: We are looking for someone to turn ideas into practical analytic solutions. The ideal candidate can bring structure to an open-ended question, determine what needs to be understood, and thoughtfully chart a clear path forward. They pair a strong analytical foundation with the ability to apply what they know and determine what makes sense for the problem at hand. They maintain momentum from the initial question through the final deliverable and care deeply about producing work that is sound and that others can understand and act on. Just as importantly, they have a track record of seeing this kind of work through successfully in a professional setting. We are looking for you to:

• Operationalize research questions from management and clients into robust analytic frameworks, research activities, and deliverables with minimal direction
• Explore and profile data sources to understand their structure, quality, and limitations; build reliable datasets and develop meaningful features for analysis and modeling
• Research and evaluate analytic methods, tools, packages, and implementation options, then select the approach best suited to the question, data, intended use, and computing environment
• Develop analyses and models using appropriate statistical, machine learning, and other analytic methods, following a disciplined, iterative process that includes clear planning, documentation, testing, and review
• Design and carry out appropriate validation, using methods such as baseline comparisons, test cases, sensitivity analyses, and record-level review
• Produce clear, well-supported deliverables, such as reports and presentations, that explain the data and methodology, interpret findings, and provide conclusions and recommendations to technical and non-technical audiences