About this role
The Analyst-Data Science joins American Express’s Credit & Fraud Risk Analytics & Data Science team to develop, deploy and validate predictive models. The role supports the use of models in economic decision-making across risk, fraud and marketing, helping the business grow profitably while managing credit losses, detecting fraud and maintaining a good customer experience.
Responsibilities
- Understand the business and the factors behind decisions across the prospect and customer lifecycle, including targeting, underwriting and customer management.
- Analyze large datasets to generate business insights and develop solutions. Use data from the American Express network to make decisions more relevant and informed.
- Develop improved approaches using big data and machine learning. Structure and communicate findings to leadership and key partners, and stay aware of developments in finance, payments and analytics.
Required qualifications
- MBA or master’s degree in economics, statistics, computer science or a related field, with 0–30 months of experience in analytics or big-data workstreams.
- Ability to work effectively in a team, deliver projects that achieve business results, communicate clearly, collaborate with cross-functional partners worldwide and work independently on complex, unstructured problems.
- Knowledge of SAS, R, Python, Hive, Spark and SQL; supervised and unsupervised techniques, including active learning, transfer learning, neural models, decision trees, reinforcement learning, graphical models, Gaussian processes and Bayesian models; and MapReduce techniques and attribute engineering.
Preferred qualifications
Expertise in coding, algorithms and high-performance computing.
The full-time role is hybrid, with locations in Gurugram and Bengaluru, India. Listed benefits include bonus incentives, financial and retirement support, insurance benefits, parental leave, wellness and counseling resources, and career development; availability of some benefits depends on location. Employment is conditional on a background verification check.