[Hiring] Credit Risk Analytics Specialist @CRNCY Group

Role Description

CRNCY Group is seeking a Credit Risk Analytics Specialist to help improve credit rule calibration and first-time loan sizing across our lending portfolio. The main objective of this role is to use historical application, loan, repayment, and collections data to determine whether our current underwriting rules are properly sizing first loans and approving the right customers. The role will focus on:

• Identifying where we may be under-lending to strong customers.
• Over-lending to higher-risk customers.
• Creating adverse selection through our current rules.

Over time, the role should help CRNCY move toward a more risk-based credit system, including:

• Stronger customer segmentation.
• Better loan amount calibration.
• Improved performance measurement.
• Risk-based pricing or variable rates.

Qualifications

• Experience helping a lender move from basic, rule-based underwriting to a more data-driven and risk-based credit model.
• Worked in environments with basic, conditional, or one-size-fits-all credit rules while maintaining strong repayment discipline, low risk tolerance, and high recovery performance.

Requirements

• Improving underwriting in practical, step-by-step stages.
• Using messy internal lending data to identify repayment patterns, customer risk, and affordability signals.
• Calibrating loan amounts based on customer risk, income, payment capacity, and repayment behavior.
• Introducing customer segmentation, scorecards, risk tiers, or probability-of-default models.
• Testing credit rule changes in controlled increments before full rollout.
• Using delinquency, default, collections, and repeat-borrowing data to improve underwriting decisions.
• Supporting the move from flat pricing or one-size-fits-all offers toward risk-based pricing or variable rates.
• Operating in markets where external credit bureau data, open banking, cashflow tools, or alternative data providers are non-existent or not fully integrated.

Technical Skills Needed

• SQL and Python to analyze application, loan, repayment, default, and collections data.
• Credit risk modeling, including probability of default, first-payment default, scorecards, and customer risk segmentation.
• First-loan sizing and affordability analysis, including payment-to-income rules and loan amount calibration.
• Modeling techniques such as logistic regression, XGBoost, LightGBM, or similar practical machine learning methods.
• Cohort analysis and portfolio performance tracking, including delinquency, default, expected loss, repeat borrowing, and collections outcomes.
• Model validation and backtesting, including out-of-time testing and data leakage prevention.
• Scenario testing and controlled experiments, including champion/challenger testing, A/B testing, Bayesian testing, causal inference, or Monte Carlo simulation.
• Predictive customer value analysis, including repeat borrowing behavior, customer lifetime value, and risk-adjusted profitability.
• Analytics and decisioning tools, such as BigQuery, Power BI, dbt, Taktile, Provenir, Alloy, Zoot, or similar platforms.

What Success Looks Like

• Clear customer risk segments.
• Better first-loan amount bands.
• Identification of under-lending pockets.
• Recommendations for rule changes and approval thresholds.
• Scenario analysis showing expected impact on approvals, defaults, collections, conversion, and profit.
• Monitoring reports to track whether changes are working.
• A practical roadmap toward risk-based pricing and scalable credit decisioning.

What We Do Not Need

We are not looking for a general business/process analyst, a research-heavy data scientist, or someone who depends on perfect external data, credit bureaus, open banking, or advanced AI tools to produce useful insights. The right person must be practical, hands-on, and able to work with the data we have today to solve the immediate first-loan sizing and underwriting calibration problem before moving into more complex modelling or long-term optimization.

Compensation

This is a contract-to-hire role with an expected hourly range of US$125–$175 per hour, depending on experience. Final compensation will be based on the candidate’s hands-on credit risk modeling experience, technical skillset, lending background, and ability to translate analysis into practical underwriting recommendations.