Staff Data Scientist (Visa Predictive Models)

Company Description

Visa is a world leader in payments and technology, with over 259 billion payments transactions flowing safely between consumers, merchants, financial institutions, and government entities in more than 200 countries and territories each year. Our mission is to connect the world through the most innovative, convenient, reliable, and secure payments network, enabling individuals, businesses, and economies to thrive while driven by a common purpose – to uplift everyone, everywhere by being the best way to pay and be paid.

Make an impact with a purpose-driven industry leader. Join us today and experience Life at Visa.

Job Description

Visa has the world’s largest consumer payment transaction dataset.  We see data on over 200 billion transactions every year from all over the world. We use that data to help our clients in the payment ecosystem grow their businesses and to help consumers access a fast, safe, and rewarding payment experience. Visa Predictive Modeling (VPM) team develops and maintains predictive machine learning models to primarily support Visa Risk and Identity Solutions. Using VisaNet data and leveraging Machine Learning (ML) and Artificial Intelligence (AI), our model scores help Visa clients all over the world for fraud defense, identity verification, smart marketing, etc. Through our models and services, VPM fuels the growth of Visa clients, generates, and diversifies revenues for VISA, while improving Visa Card customer experience and their financial lives.

Within VPM, the Authentication Model Development Team is responsible for developing real-time fraud detection models during card transaction’s EMV 3-D Secure (3DS) Authentication process. We leverage a set of rich data available during the 3DS authentication including transactional, digital and identity information to detect and stop fraud.

This is a Technical role. Your responsibilities include :

• Building and validating predictive models with advanced machine learning techniques and tools to drive business value, interpreting and presenting modeling and analytical results to non-technical audience.

• Conducting research using latest and emerging modeling technologies and tools (e.g., Deep Neural Networks, RNN, LSTM, etc.) to solve new fraud detection business problems like enumeration attacks and/or improve existing production models’ performance.

• Improving the modeling process through MLOps and automation to drive efficiency and effectiveness.

• Partnering with a cross functional team of Product Managers, Data Engineers, Software Engineers, and Platform Engineers to deploy models and/or model innovations into production.

• Managing model risks in line with Visa Model Risk Management requirements.

• Conducting modeling analysis to address internal and external clients’ questions and requests.

This is a hybrid position. Hybrid employees can alternate time between both remote and office. Employees in hybrid roles are expected to work from the office 2-3 set days a week (determined by leadership/site), with a general guidepost of being in the office 50% or more of the time based on business needs.

Qualifications

Basic Qualifications
• 5 or more years of relevant work experience with a Bachelors Degree or at least 2 years of work experience with an Advanced degree (e.g. Masters, MBA, JD, MD) or 0 years of work experience with a PhD

Preferred Qualifications
• 6 or more years of work experience with a Bachelors Degree or 4 or more years of relevant experience with an Advanced Degree (e.g. Masters, MBA, JD, MD) or up to 3 years of relevant experience with a PhD in a quantitative or engineering field
• Prior experience with production model development, implementation, and support.
• Prior experience with modern machine learning framework and tools like Gradient Boosting (e.g., XGBoost)
• Experience with Payment Fraud models.
• 3 or more years of experience with production model development, implementation, and support.
• Experience with modern Deep Neural Network framework (e.g. RNN, LSTM, Transformer) and tools (e.g., PyTorch, TensorFlow).
• Experience in Agile Development and tools.
• Proven ability to quickly learn and apply new tools and techniques.
• A strong innovation leader yet a practitioner to create tangible business value balancing business objectives and technological constraints.
• Must be a team-player and capable of handling multi-tasks in a dynamic environment.
• Excellent business writing, verbal communication, and presentation skills to technical as well non-technical audiences.

Technical Qualifications
• Proficiency in Python, Hadoop, Hive, Spark for big data analysis and modeling
• Experience