Risk Program Senior Associate

What you'll do

• Identify and retool machine learning algorithms to detect fraud within the Chase Consumer Bank.
• Perform feature engineering and feature selection on datasets containing billions of transactions.
• Develop and train machine learning algorithms using cutting-edge technology to extract predictive patterns.
• Utilize big data platforms and analytical capabilities to control fraud loss and improve customer experience.
• Provide guidance to business teams on leveraging machine learning solutions to meet specific needs.
• Share knowledge and disseminate findings regarding machine learning models to a broad internal audience.
• Design and implement graph-based machine learning models for risk assessment and fraud detection.
• Fine-tune and deploy transformer-based models using frameworks like PyTorch or TensorFlow.

What we're looking for

• Master's degree in Mathematics, Statistics, Economics, Computer Science, Operations Research, Physics, or other related quantitative fields.
• 2+ years of experience with data analysis in Python.
• Experience designing models for a commercial purpose using at least three of the following: CNN/RNN, Reinforcement Learning, Random Forest/GBM, Transformer based architecture, or Graph technology.
• A strong interest in how models work, why they succeed or fail on specific problems, and the practical aspects of new model design.
• PhD in a quantitative field with publications in top journals, preferably in machine learning (preferred).
• Proven experience designing models in cloud environments and utilizing distributed/parallel processing across Databricks, AWS, and GCP (preferred).
• Experience with graph technology, including designing and implementing graph-based machine learning models for fraud detection or risk assessment (preferred).
• Hands-on experience with transformer models and related architectures for NLP, anomaly detection, or transaction analysis (preferred).