Lead/Senior Data Scientist, Cleo AI

• 4+ years analytics or data science experience in a risk-focussed role, ideally within fintech, lending, or payments
• Excellent SQL skills
• Fluency in Python (or R) for data analysis, modelling, and statistical testing
• Experience conducting large scale A/B experiments and interpreting results to drive product and business decisions
• Fluent in credit portfolio metrics – e.g. arrears buckets, roll rates, loss rate, yield/marginal loss – and how they tie to unit economics and P&L
• Hands‑on experience working with predictive models (e.g. credit, fraud, marketing), including interpreting metrics like AUC/Gini, calibration, PSI/CSI, drift.Hands-on experience with BI tools (e.g. Looker, Mode, Tableau) and data workflow tools (dbt, Airflow)Strong analytical rigour and the ability to translate findings into clear business recommendationsTrack record of taking analyses all the way through to shipped changes and measurable impact

Desirable

• We’re looking for a Lead/Senior Data Scientist to help us measure, monitor, and improve the performance of Cleo’s credit products. This is a level 3 or 4 Analytics role
• This is a hands‑on data science and analytics role. You’ll be analysing behaviour across millions of US users, using rich transactional and behavioural that powers Cleo’s AI money coach and credit products
• You’ll spend the majority of your time in SQL and Python, working directly from Cleo’s data warehouse to understand, explain, and improve credit performance
• This is not a traditional underwriting or policy role. You’ll be the analytics owner for [EWA / specific product], with direct line of sight to losses, revenue, and product roadmap
• You’ll work closely with other analysts, Risk Modellers, Product Managers, and Engineers to diagnose portfolio trends, build monitoring frameworks, and deliver insights that inform how Cleo manages and optimises risk
• You’ll sit within the Risk & Payments pillar, working at the intersection of data, decisioning, and product, helping us build scalable systems that balance user access with sustainable economics
• You’ll be part of a growing team responsible for driving profitable growth while protecting the business from loss, using data to understand repayment behaviour, model performance, and system-level trade-offs
• This is an opportunity to shape how we quantify and manage risk as we expand across new credit products and geographies
• 1. Credit & Risk Performance Analytics:
• Write complex SQL/python to pull cohort‑ and event‑level datasets from our warehouse and turn them into clear, decision‑ready analyses
• Quantify the commercial impact of performance changes (losses, yield, approval rate)
• Design and analyse multivariate experiments on underwriting, pricing, or repayment flows, and translate results into actionable risk strategies
• Analyse arrears, default, and yield trends across Cleo’s credit products
• Identify emerging risks and shifts in eligibility or repayment behaviour using cohort and segmentation analysis
• Build and maintain dashboards for portfolio health and performance tracking
• Design early-warning alerts for anomalies in arrears or model-driven decisioning
• 2. Model Understanding & Monitoring:
• Partner with the Risk Modelling team to turn model health metrics (AUC, PSI, calibration, feature drift) into clear recommendations for policy or product changes
• Monitor model stability and support investigations into concept drift and feature degradation
• Quantify the impact of model changes and assess whether observed shifts are model- or market-driven
• 3. Deep-Dive Investigations:
• Conduct root-cause analysis on performance deteriorations (e.g., arrears spikes, yield compression)
• Own investigations from question → analysis → recommendation, and present your work to Risk, Product, and Leadership
• Use decomposition, SHAP analysis, and driver frameworks to explain variance in loss and yield
• Support the design and measurement of A/B tests or pilot changes in credit decisioning or repayment operations
• 4. Forecasting & Scenario Support:
• Partner with Finance and Commercial teams to support variance analysis and monthly forecast inputs
• Model how shifts in repayment or eligibility rates flow through to portfolio loss and profitability
• 5. Tooling, Frameworks & Collaboration:
• Work with Analytics Engineering to improve risk data pipelines and metric definitions
• Build reusable analysis templates and frameworks for monitoring across multiple credit products
• Communicate insights clearly to non-technical stakeholders, transforming complex findings into actionable decisions