Job Summary

As a Data Scientist at BNIC you will design, build and operationalize predictive models and analytics that drive telco business decisions (e.g., churn, ARPU, sales targets, network investment). You will work end-to-end—from data preparation and exploratory analysis to model deployment and monitoring—collaborating closely with product owners, data engineers, MLOps and business stakeholders.

Job Responsibilities*

Data Preparation & Exploration

• Collect, clean and prepare telecom datasets (subscriber records, usage, billing, sale, and network KPIs).
• Perform exploratory data analysis (EDA) to identify patterns, drivers and features.

Modeling & Feature Engineering

• Develop and validate ML models (classification, regression, clustering) for churn, sales/ARPU prediction, KPI forecasting and anomaly detection.
• Implement feature engineering, data validation checks and reproducible training pipelines.

Deployment & Governance

• Deploy models and pipelines to production using MLOps best practices; implement monitoring and retraining flows.
• Lead model governance activities: interpretability, bias checks, performance monitoring and documentation.

Communication & Business Alignment

• Create dashboards and presentations to communicate insights and business impact to stakeholders and executives.
• Translate business objectives into measurable ML success criteria and collaborate to prioritize work.

Qualifications

•    Bachelor’s or Master’s degree in Statistics, Computer Science, Data Science, Engineering or related field.
•    3–5 years professional experience building and deploying ML models (telecom experience preferred).

•    Strong Python skills (pandas, scikit-learn, XGBoost), SQL and experience producing reproducible analysis (notebooks, scripts, version control).
•    Familiarity with cloud ML platforms and MLOps tools (e.g., AWS SageMaker, GCP Vertex AI, MLflow, CI/CD).
•    Experience with data viz tools (Power BI, Tableau) and communicating results to non-technical stakeholders.
•    Strong problem solving and statistical foundations (A/B testing, evaluation metrics, time-series basics).