Data Scientist

Broadgate; United States; On-site; USD 120,000 - 180,000; Full time

Job summary

Broadgate in the United States is seeking a client-facing data scientist who will build credit models for prospective lending clients to quantify how risk signals improve approvals and reduce losses. You will run experiments, deliver results to clients, and ensure models meet US fair lending standards.

You'll also contribute to product roadmaps by translating insights into actionable recommendations for both technical and commercial audiences.

Qualifications

• Experience taking raw data to production-quality ML models.
• Ability to design features and measure performance.
• Comfort presenting results to technical and business audiences.

Responsibilities

• Work with prospective clients before they sign, building tailored models on their data to measure lift from risk signals.
• Show clients the results directly, explaining improvements in approvals and losses.
• Run tests and experiments to validate new features and modelling approaches.
• Ensure models and analyses meet US fair lending standards and incorporate these into feature design.
• Stay current on developments in ML, AI and credit risk tooling.

Skills

Python

scikit-learn

XGBoost

Data exploration

Model production

Tools

XGBoost

Gradient boosting

Job description

This is a client-facing data science role in the Data Science team, reporting to its leader. You’ll build credit models with prospective lending clients to show how much our alternative-data risk signals improve their results. The main measures are more approvals and lower losses. What you learn from these engagements will also shape how our data products and roadmap develop.

What You’ll Do

• Work with prospective clients before they sign, building tailored models on their data to measure the lift from our risk scores and features.
• Show clients the results directly, explaining what the improvements in approval rates and credit losses mean for their business.
• Run tests and experiments to check how well new features and modelling approaches work.
• Make sure models and analyses meet US fair lending standards, and build those requirements into how features are designed.
• Stay current on new developments in machine learning, AI and credit risk tooling.

What You’ll Bring

• A background in US consumer or small-business lending, fintech, or financial services.
• A track record of taking raw data all the way to production-quality ML models, including feature design and performance measurement.
• Hands-on skill with Python ML tools, especially scikit-learn and gradient-boosted tree methods (XGBoost or similar).
• Strong instincts for exploring data, and the ability to find what matters and turn it into clear recommendations.
• Confidence presenting to both technical and commercial audiences, including external clients.
• A methodical, detail-focused approach to solving problems.