As a Sr Data / ML Scientist you will help design, enhance, and build next generation fraud detection solutions in an agile development environment.
Formulate business problems as technical data problems while ensuring key business drivers are captured in collaboration with product stakeholders.
Work with software engineers to ensure feasibility of solutions. Deliver prototypes and production code based on need.
Experiment with in-house and third-party data sets to test hypotheses on relevance and value of data to business problems.
Build needed data transformations on structured and un-structured data.
Build and experiment with modeling and scoring algorithms. This includes development of custom algorithms as well as use of packaged tools based on machine learning, data mining and statistical techniques.
Devise and implement methods for adaptive learning with controls on effectiveness, methods for explaining model decisions where necessary, model validation, A/B testing of models.
Devise and implement methods for efficiently monitoring model effectiveness and performance in production.
Devise and implement methods for automation of all parts of the predictive pipeline to minimize labor in development and production.
Contribute to development and adoption of shared predictive analytics infrastructure.
Mentor and train other ML scientists on the team on key solutions
Able to work on multiple projects and initiatives with different/competing timelines and demands.
Present technical solutions, capabilities, considerations, and features in business terms. Effectively communicate status, issues, and risks in a precise and timely manner
Collaborate across engineering teams and leaders in VERA, Visa Research, AI Platform, Operations, and Infrastructure (O&I), security and platform teams.
This is a hybrid position. Expectation of days in office will be confirmed by your hiring manager.
8 or more years of relevant work experience with a Bachelor Degree or 7 or more relevant years of experience with an Advanced Degree (e.g. Masters, MBA, JD, MD) or 3 or more years of experience with a PhD
Preferred Qualification
Expert in leading-edge areas such as Machine Learning, Deep Learning, Stream Computing and MLOps
High level of competence in Python, Perl, Java, Scala, and/or Unix/Linux scripts highly preferred
Extensive experience with SAS/SQL/Hive for extracting and aggregating dat
Experience with Big Data and analytics leveraging technologies like Hadoop, Spark, Scala, and MapReduce
Deep learning experience working with TensorFlow and Natural Language
Processing experience are highly preferred.
Experience with one or more common statistical tools such SAS, R, KNIME, Matlab
Experience in developing large scale, enterprise class distributed systems of
high availability, low latency, & strong data consistency
Experience in architecting solutions with Continuous Integration and Continuous Delivery in mind
Familiarity with in distributed in-memory computing technologies