← Back to Data & Analytics All sample CVs →

Sample document — fictional candidate for illustration purposes only.

Tariq Osei
Senior Data Scientist
tariq.osei@email.com +31 6 12 345 678 Amsterdam, Netherlands github.com/tariqosei linkedin.com/in/tariqosei
Technical Stack
Python R SQL PySpark scikit-learn PyTorch XGBoost dbt Airflow Databricks BigQuery Looker Tableau Azure ML
Professional Summary

Senior Data Scientist with 5 years building and deploying production ML systems in fintech and e-commerce. Shipped a fraud detection model reducing false positives by 38% at €2.1M annual saving. Built churn prediction pipeline serving 4.8M customers. Strong in translating ambiguous business problems into reproducible, production-ready analytical solutions.

Experience
Senior Data Scientist Mar 2022 – Present
Adyen N.V., Amsterdam
  • Business question: Reduce card fraud without increasing false positive rate; Approach: Rebuilt fraud scoring model using gradient boosting with 140 engineered features on 18-month transaction history; Outcome: False positives down 38%, fraud losses down 22%, saving €2.1M annually.
  • Built real-time feature store in Python + Redis serving model inference at p99 < 8ms at 14,000 TPS.
  • Designed A/B testing framework adopted by 3 other data science teams; enabled 40% faster experiment cycles.
  • Mentored 2 junior data scientists; both promoted to mid-level within 14 months.
Data Scientist Jun 2020 – Feb 2022
Jumia Technologies, Nairobi (Remote)
  • Business question: Identify customers likely to churn within 90 days; Approach: Survival analysis + logistic regression on 4.8M customer base; Outcome: Retention campaign ROI of 3.4x in first quarter of deployment.
  • Built demand forecasting model for 12 product categories; reduced overstock by 18% and stockouts by 24%.
  • Established first data science notebook standards and reproducibility guidelines for a team of 8 analysts.
Data Analyst Jan 2019 – May 2020
Equity Bank, Nairobi
  • Built credit risk dashboards in Tableau used daily by 14 credit officers; reduced manual reporting by 6 hours/week.
  • Developed SQL-based pipeline for regulatory reporting, cutting monthly close time from 3 days to 6 hours.
Projects & Publications
  • afridata-benchmark — open dataset of African e-commerce transaction patterns (GitHub, 890 stars)
  • "Real-time fraud detection at scale: lessons from fintech" — PyData Amsterdam 2025 (talk)
  • Kaggle Competition — Tabular Playground: top 4% finish, 2024
Education
M.Sc. Data Science Sep 2017 – Jun 2019
University of Cape Town — Distinction; Thesis: Predictive modelling for mobile money fraud in sub-Saharan Africa
B.Sc. Statistics & Mathematics Sep 2013 – Jun 2017
University of Ghana — First Class Honours