Lead Quantitative Analytics Specialist – AIM Fraud Modeling

Wells Fargo Bank•Charlotte, NC

7d•Hybrid

About The Position

Wells Fargo is seeking a Lead Quantitative Analytics Specialist within its AI Innovation Modeling (AIM) group under COO Data, Analytics & AI. The role focuses on developing, validating, and advancing machine learning and statistical models to detect and prevent fraud. The specialist will work with large-scale transaction and application data to identify emerging fraud patterns, engineer predictive features, assess sophisticated models, and improve performance from research through production implementation. The position combines hands-on model development with independent validation, effective challenge, documentation, governance, and collaboration across fraud, technology, model risk, audit, and business teams. The specialist will help apply machine learning, AI, and emerging GenAI capabilities responsibly to fraud detection and prevention, while communicating findings to technical and non-technical audiences and supporting customer protection and business outcomes. The position is based in Charlotte, North Carolina, and follows a hybrid schedule of three days on-site and two days remote. Remote work locations are not available. The role may require travel up to 10% of the time and is not eligible for visa sponsorship.

Requirements

At least 5 years of quantitative analytics experience, or equivalent experience demonstrated through work, training, military experience, or education.
Bachelor’s degree or higher in a quantitative discipline such as mathematics, statistics, engineering, physics, economics, or computer science.
The ability to work on-site according to Wells Fargo’s hybrid schedule of three days on-site and two days remote.
Ability to travel up to 10% of the time.
The position is not eligible for visa sponsorship.

Nice To Haves

Master’s degree or higher in a quantitative discipline.
At least 3 years of fraud detection modeling or analytical experience in banking or financial services, including hands-on machine learning model development and/or independent model validation.
Experience developing, validating, or challenging fraud models across the model lifecycle, including data assessment, sampling, feature engineering, model design and specification, performance testing, implementation review, documentation, and ongoing monitoring.
Hands-on experience with supervised and unsupervised learning techniques such as CNN, RNN, LSTM, AutoEncoder, SVM, reinforcement learning, isolation forest, random forest, XGBoost, or GBM.
Experience evaluating fraud models using methods suited to imbalanced classification, including discrimination, calibration, stability, threshold selection, false-positive trade-offs, and out-of-time performance.
Experience with NLP, LLMs, and GenAI, including assessing appropriate use cases, model risks, limitations, and controls.
Experience developing models with Keras, TensorFlow, or PyTorch.
Experience designing and executing models in big data environments using Hadoop, Spark, or distributed processing frameworks.
Advanced data analysis and wrangling skills using Python, pandas, PySpark, NumPy, SQL, or similar tools.
Ability to translate complex quantitative findings into clear recommendations for business leaders, model governance partners, auditors, and regulators.
Strong technical documentation, problem-solving, collaboration, and communication skills, with a detail-oriented and results-driven approach to high-quality work.

Responsibilities

Lead end-to-end development and enhancement of machine learning and statistical models for real-time fraud risk assessment, transaction or application ranking, and fraud prevention.
Analyze millions to billions of records in big data environments to identify predictive fraud signals, evolving attack patterns, and opportunities to reduce fraud losses while supporting customer experience.
Build Python-based modeling pipelines covering data preparation, sampling, feature engineering, feature selection, model training, testing, and performance evaluation.
Apply and assess advanced modeling techniques, including gradient boosting, random forests, isolation forests, support vector machines, neural networks, autoencoders, and sequence models.
Perform independent model validation and effective challenge, evaluating conceptual soundness, data quality, assumptions, methodology, implementation, performance, stability, limitations, and ongoing monitoring.
Design model evaluation strategies using appropriate fraud measures, threshold analysis, out-of-time testing, back-testing, sensitivity analysis, and benchmarking.
Lead debugging, testing, code review, and performance tuning for models developed in Python and distributed computing environments.
Document model methodology, validation evidence, assumptions, limitations, controls, and performance outcomes in a clear, reproducible, governance-ready manner.
Present analytical findings and recommendations to technical and non-technical audiences, including fraud leaders, business stakeholders, model validation, model risk, internal audit, governance partners, and regulatory reviewers.
Partner with business, technology, data, and risk teams to move analytically sound fraud solutions into production and ensure models address changing fraud threats.
Mentor or provide technical leadership to quantitative analysts and contribute to standards, reusable modeling practices, and analytical rigor.
Perform targeted analysis and reporting in response to emerging fraud trends, model-performance questions, audit requests, or regulatory review needs.