Data Scientist II -Card Fraud Analytics

Truist•Atlanta, GA

1d•Onsite

About The Position

The Data Scientist II position involves performing sophisticated analytics, including statistical and predictive analytics and machine learning modeling, to generate actionable insights that enhance business outcomes and mitigate risk. The role also entails consulting business leaders and stakeholders on leveraging these insights to develop analytical strategies. Specifically, this role supports the fraud strategy function by developing and refining fraud rules, conducting analyses to identify and retire underperforming strategies, and participating in Card Fraud benchmarking forums. The teammate will apply advanced analytics techniques, including machine learning fraud rule generation, to refit top fraud rules and work with geo-location datasets to identify fraud patterns.

Requirements

Bachelor’s degree and four or more years of experience in a quantitative field such as Finance, Mathematics, Analytics, Data Science, Computer Science, or Engineering, or equivalent education and related training.
Exhibit understanding of statistical methods, including a broad understanding of classical statistics, probability theory, econometrics, time-series, and primary statistical tests.
Familiarity with linear algebra concepts for optimization, complex matrix operations, eigenvalue decompositions, and principal components; working knowledge of calculus/differential equations, with understanding of stochastic processes.
Demonstrate understanding of data cleansing and preparation methodologies, including regex, filtering, indexing, interpolation, and outlier treatment.
Strong familiarity with data extraction in a variety of environments (SQL, JQuery, etc.).
Working knowledge of Hadoop, Pig, Hive, and/or NoSQL, Spark.
Experience in managing multiple projects with tight deadlines in a collaborative environment.
Working knowledge and/or hands-on fraud detection experience with card products, card processors, card networks, and one or more fraud rule systems such as Defense Edge, FDWC, Falcon Expert, Tsys Card Guard, Tsys Determinator, Broadcom 3DS, Token Administration, Visa Risk Manager.
Experience with working with geo-location datasets used to identify patterns of fraud (online and/or offline).

Nice To Haves

Master’s degree or PhD in a quantitative field such as Finance, Mathematics, Analytics, Data Science, Computer Science, or Engineering.
Four years of relevant work experience if candidate lacks graduate degree.
Previous experience in the banking or fin-tech industry.

Responsibilities

Independently perform sophisticated data analytics (ranging from classical econometrics to machine learning, neural networks, and natural language processing) in a variety of environments using structured and unstructured data, specifically tailored to fraud problems.
Produce compelling data visualizations to communicate insights and influence outcomes among a wide array of stakeholders.
Take accountability and ownership of end-to-end data science solution design, technical delivery, and measurable business outcomes.
Engage in stakeholder meetings to identify business objectives and scope solution requirements.
With minimal guidance, write, document, and deploy custom code in a variety of environments (Python, SAS, R, etc.) to create predictive analytics applications.
Use, maintain, share and collaborate through Truist internal code repositories to foster continual learning and cross-pollination of skillsets.
Actively research and advocate adoption of emerging methods and technologies in the data science field, with the eye of continually advancing Truist’s capabilities.
Exercise sound judgment and foster risk management culture throughout design, development, and deployment practices; partner with cross-functional teams to coordinate rules on data usage, data governance and analytics capabilities.
Support the fraud strategy function by developing 3 or more fraud rules per month.
Conduct analyses to identify underperforming fraud strategies that need to be retired.
Participant at Card Fraud benchmarking forums with deep industry knowledge to help our team advance with modern fraud strategies.
Capability to develop or leverage the advanced analytics techniques of machine learning fraud rule generation to refit our top 20% of fraud rules per fraud rule platform.

Benefits

medical
dental
vision
life insurance
disability
accidental death and dismemberment
tax-preferred savings accounts
401k plan
no less than 10 days of vacation (prorated based on date of hire and by full-time or part-time status) during their first year of employment
10 sick days (also prorated)
paid holidays
defined benefit pension plan (depending on the position and division)
restricted stock units (depending on the position and division)
deferred compensation plan (depending on the position and division)