Sr Data Scientist - Fraud
- Career Level
- Not Specified
The primary objective of this position is to ensure that Data Analytics COE department delivers fraud analytics and provide data-driven, action-oriented solutions to business problems through statistical data mining, analytics techniques and a consultative approach to the Enterprise. This Sr. Data Scientist will be a core member of the Data Analytics team and will play a significant role in helping Assurant advance and create data driven fraud analytics solutions. In this role, the Sr. Data Scientist incorporates techniques across many disciplines – including mathematics/statistics, computer programming, data engineering, data management, visualization and ETL. The incumbent will support fraud analytics product development and high performance computing with traditional business expertise with the goal of using data to optimize risk and fraud related business decisions. Also, the incumbent will act as an evangelist for data science and be an expert/fluent in several of these data science disciplines; sufficiently proficient in others to effectively design, build, and deliver end-to-end fraud predictive analytics solutions/products to optimize business decisions; will enjoy working with some of the most diverse global data sets, cutting edge technology, and the ability to see data insights turned into real business results on a regular basis. This role requires to partner with leaders in various divisions, clients and geographies, in order to ensure that increasingly more data driven solutions are brought to the Data Analytics group. This individual will support fraud solutions, products and services across all line of businesses within Assurant and an in-depth understanding of data, database management systems, statistics, predictive modeling, and machine learning is required.
40% - Lay the groundwork – hypothesize as an individual researcher and in collaboration with other team members on how to solve problems. Perform data preparation activities, such as collecting, cleaning, and organizing.
• Analyze effectiveness of fraud models to constantly improve tools, procedures, and workflows that minimize risk and enhance customer experience
• Uses best practices to understand the data and develop statistical, machine learning techniques to build models that address business needs.
• Collaborates with the team in order to improve the effectiveness of business decisions through the use of data and machine learning/predictive modeling.
• Understands the business' problems to identify the optimal business solution/modeling approach and support your answers and findings with appropriate statistical techniques and methods
30 % - Turn data into insight – segment, cluster, model, and mine to better understand the behavior in question. Explain what has happened or predict what
will in an actionable fashion.
• Transform data into insights, to identify and quantify opportunities to reduce fraud and false positive into a positive business impact.
• Use and leveraging internal and external Fraud tools as part of our Fraud operations (e.g., R, SAS, Python, SQL, Hadoop)
30% - Drive change – produce clear, understandable visualizations and reports to share with Senior Management. Partner with product, digital, engineering, marketing and all line of business to design tests and implement your model findings insights.
• Communicates to team members, leadership and stakeholders on findings to ensure models are well understood and incorporated into business processes.
• Manages data and data requests to improve the accuracy of our data and decisions made from data analysis.
• Participate and drive data modeling and governance best practices.
• 5 years of relevant experience in risk & fraud analytics/operations.
• Master Degree in a quantitative field, such as Data Analytics, Statistics, Mathematics, Computer Science, Finance.
• 5 years of relevant experience in analytics, statistical/quantitative modeling and/or machine Learning tools (R, Python, etc.) and in using various database tools (e.g. Hadoop, SQL) processing large volumes of structured and unstructured data.
• Strong understanding of risk and fraud management in insurance or financial environment required.
• Experience in managing and manipulating large, complex datasets and techniques to build models that have driven company decision making.
• Experience in working with any or all of statistical/data processing software such as R, SAS, Weka, SPSS, MatLab, CART etc.
• Experience in database such as SQL, Hadoop, NoSQL, Massively Parallel Processing (MPP) databases.
• Proficient and ability to code and develop prototypes in programming languages in Python, Java, Perl, JEE, .Net, C#, or C++.
• Ability to work with unstructured data, whether it is from digital, social media, video feeds or audio, device logs, etc.
• Advocate machine learning principles to become a SME within the organization.
• Organized and capable of independently managing complex analytical projects from start to finish.
• Ability to independently structure analyses data; interpret moderate to complex analytical concepts/models and communicate the findings to a non-technical audience.
• Ability to provide written and oral interpretation of highly specialized terms and data, and ability to present this data to others with different levels of expertise.
• Manage stakeholder relationship
• Demonstrated analytic agility.
• Understanding of the Insurance industry/market place and regulation preferred