Data Science Support Intern at Pezesha

The Data Team Support Intern will support both Data Scientists and Data Engineers in testing, validating, and monitoring credit scoring and transaction classification systems.
The role is designed as a hands-on learning position, where the intern will assist with model testing, feature validation, ETL checks, deployment verification, and reporting, while gaining practical exposure to machine learning models, data pipelines, and production systems.


Key Responsibilities


Model Testing & Validation (Data Science Support)
M-Pesa Classifier Testing
ETL & Deployment Support (Data Engineering Support)
Documentation & Change Tracking


Required Qualifications


Assist in testing credit scoring models built by Data Scientists.
Help verify the accuracy and consistency of features used in models.
Support basic validation of model outputs across test and production environments.
Participate in regression testing when models or features are updated.
Support testing of the M-Pesa transaction classifier, including LLM-based classification logic.
Assist in measuring classification accuracy and identifying misclassified transactions.
Help track and report classifier accuracy, with a target of above 95% under supervision.
Assist in validating ETL pipelines that feed credit scoring systems.
Help check data accuracy, completeness, and consistency in processed datasets.
Support testing of deployed models to ensure they run correctly after release.


Assist in documenting:


Model and feature changes
ETL updates
Testing results and observations
Maintain simple change logs and testing notes for internal reference.


Currently pursuing or recently completed a degree in:


Data Science
Computer Science
Statistics
Information Technology
Or a related field


Basic knowledge of:


Python and SQL
Strong R programming skills
Data analysis concepts
Machine learning fundamentals