This leader will design and scale a fully auditable, real-time, data-driven lending engine capable of operating in emerging markets with initially minimal data and gradually expanding to rich alternative-data environments.
Key Responsibilities
Risk & Decision Science Leadership
- Build end-to-end credit-decision engines for thin-file and micro-loan lending.
- Develop dynamic risk-based pricing, approval strategies, and behavioural scorecards.
- Design and maintain real-time PD/LGD models, portfolio-risk dashboards, and EWS triggers.
- Manage the full model-risk governance cycle, including documentation and audit trails.
- Align Product, Engineering, Data and Collections teams around unified risk limits and target ROE.
- Oversee portfolio monitoring, NPL caps, loss forecasting and scenario modelling.
Data & Architecture Ownership
- Define and deliver the company’s real-time data architecture:
- Streaming ELT data lake/warehouse feature store ML-ops pipeline
- Make strategic architectural choices (e.g., Kafka vs. Kinesis, Delta vs. Iceberg).
- Ensure robust data quality, lineage and metadata management using tools such as:
- Great Expectations, DataHub, Collibra
- Build and scale company-wide BI, reporting, KPI frameworks and data literacy programs.
- Support engineering in building scalable, compliant microservices for credit and data operations.
Leadership & Team Management
- Build and lead multi-disciplinary teams: risk analysts, data engineers, ML engineers, BI analysts.
- Mentor future leaders and drive a high-performance, safety-first, analytically rigorous culture.
- Ensure strong cross-functional collaboration across Product, Engineering, Finance, Collections and Operations.
Requirements
Experience
- Hand-on experience building credit-risk engines, pricing models and data pipelines for thin-file or micro-loan lending.
- Ability to work initially with minimal data (phone + ID only) and scale into rich alt-data ecosystems (device, telco, behavioural, psychometric, open banking).
- Commercial instinct to balance conversion rate, data cost and decision quality, managing approval-rate vs. portfolio-yield trade-offs.
- Experience in regulated environments with a clean compliance reputation.
- Hand-on coding in Python or R, strong SQL, and understanding of ML-ops latency/production constraints.
- Demonstrated experience presenting frameworks to regulators and auditors.
Skills & Competencies
- Deep understanding of portfolio management: vintage curves, NPL, ROE, collections strategy.
- Strong architectural judgement: streaming, storage formats, orchestration, ML-ops.
- Ability to link dashboards, models and KPIs directly to commercial OKRs.
- Strong cross-functional communication and executive-level storytelling.
- Ability to build scalable teams in fast-moving, ambiguous environments.
Mindset
Hand-on, pragmatic, commercially minded.
Scientific but not academic — every model tied to P&L.
Zero tolerance for compliance shortcuts.
High ownership, high integrity, high urgency.
Hand-on experience building credit-risk engines, pricing models and data pipelines for thin-file or micro-loan lending.
Ability to work initially with minimal data (phone + ID only) and scale into rich alt-data ecosystems (device, telco, behavioural, psychometric, open banking).
Commercial instinct to balance conversion rate, data cost and decision quality, managing approval-rate vs. portfolio-yield trade-offs.
Experience in regulated environments with a clean compliance reputation.
Hand-on coding in Python or R, strong SQL, and understanding of ML-ops latency/production constraints.
Demonstrated experience presenting frameworks to regulators and auditors.
