Skip to content

JD — Data Scientist

Owner Classification Review Date Status
People Operations Internal April 2027 Active

Job Description: Data Scientist

Department: Technology & Digital — Data
Reports to: Head of Data

Role Overview

Simpaisa processes over $1 billion annually across 7 markets. The question "where is the money, why didn't it arrive, and which operator is the problem?" is answered manually today. This role exists to change that.

The Data Scientist will build the analytical and machine learning foundation that transforms Simpaisa's transaction data into operational intelligence. The immediate priorities are: CCQ analytics (Coverage, Cost, Quality across all corridors), reconciliation automation, routing optimisation, and fraud signal detection. Longer term, this role contributes to the AI/ML platform that will power real-time decisioning across the payment network.

This is an applied role in a production fintech environment — not research. Models are valuable when they run in production and save someone 4 hours of manual reconciliation. That is the bar.

Key Responsibilities

  • Design, build, and deploy machine learning models and statistical analyses for payment operations — corridor performance, operator quality, reconciliation anomalies, fraud signals.

  • Build the CCQ (Coverage, Cost, Quality) analytics layer: turn transaction state data into operator scorecards, corridor health dashboards, and SLA tracking.

  • Develop reconciliation automation: identify unmatched transactions, classify discrepancy types, and reduce manual exception handling.

  • Build routing optimisation models: given CCQ data, predict the highest-quality, lowest-cost operator for each transaction at dispatch time.

  • Work with Data Engineers to productionise models — feature engineering, model serving, monitoring, drift detection.

  • Communicate findings to non-technical stakeholders (CSNO, CPO, Head of Finance) through clear visualisations and narrative.

  • Contribute to the AI/ML platform design — MLflow, feature stores, model registry — as the data team scales.

  • Stay current with techniques relevant to financial services: anomaly detection, time-series forecasting, graph-based fraud detection.

Required Skills and Experience

  • Python: Proficiency in Python for data analysis and model development. Pandas, NumPy, scikit-learn as baseline; PyTorch or TensorFlow for deep learning work.

  • ML fundamentals: Strong understanding of supervised and unsupervised learning, model evaluation, cross-validation, and common failure modes (overfitting, data leakage).

  • SQL: Comfortable writing complex queries for analysis and feature engineering on transactional databases.

  • Statistical analysis: Time-series analysis, anomaly detection, A/B testing, significance testing.

  • Production ML: Experience deploying models to production — not just notebooks. Understanding of model monitoring and retraining cycles.

  • Data visualisation: Ability to build dashboards (Metabase, Grafana, Tableau, or similar) and present insights clearly.

  • Payments domain (preferred): Experience with payment data, transaction logs, settlement records, or reconciliation systems is a strong advantage.

  • Communication: Ability to translate a model's output into a business recommendation. "The model says" is not enough.

General Requirements

  • Bachelor's degree in Statistics, Mathematics, Computer Science, Economics, or a related quantitative field. Advanced degree preferred.

  • 5+ years of experience in data science or analytical roles with a focus on production model delivery.

  • Demonstrated track record of building models that ran in production and delivered measurable business value.

What We Offer

  • Competitive salary benchmarked to your local market.

  • Transaction data at a scale and complexity few data scientists encounter — 7 markets, multiple currencies, hundreds of operator APIs.

  • Clear mandate: initiative 3.4 in the CDO 30/60/90 plan is yours to build.

  • Direct access to CDO and CSNO — your outputs will inform corridor strategy.

  • Flexible hybrid working.