Data Scientist at Taraki

Taraki is hiring for one of its clients.Location: RemoteExperience Level: 5 to 8 years in Applied Machine Learning, Statistical Modeling, and Data Science for large-scale systems Department: Data AI EngineeringCompensation: PKR 600,000 to 850,000 (based on experience)Role Summary:The Data Scientist will develop ML models to predict customer behavior, price sensitivity, and offer conversion probability. They will build and maintain elasticity models, segmentation models, and reinforcement learning frameworks to enable personalized pricing and offer recommendations.Key ResponsibilitiesBuild and deploy models for:Price Elasticity / Conversion PredictionChurn Propensity / Retention UpliftSegment Discovery Similarity (Clustering, KNN)Offer Recommendation / Ranking (Scoring Models)Design A/B testing and uplift modeling to evaluate campaign performance.Develop simulation engines for pricing what-if analysis and scenario testing.Create automated pipelines for model training, scoring, and retraining.Work closely with Data Engineers to ensure feature store alignment.Collaborate with the Business Decisioning team to translate insights into rules and thresholds.Implement feedback loops using real-time events (purchase, rejection, expiry) to improve models.Required SkillsStrong foundation in Machine Learning, Statistics, and Econometrics.Proficient in Python (pandas, scikit-learn, numpy, statsmodels, xgboost, lightGBM).Experience with model lifecycle management (MLOps).Solid understanding of telecom KPIs: ARPU, recharge frequency, wallet size, churn rate, etc.Ability to design feature engineering pipelines and perform A/B testing.Expertise in data visualization and storytelling for non-technical stakeholders.Tools TechnologiesData Science: Python, R, Scikit-learn, XGBoost, LightGBM, TensorFlow, PyTorchMLOps: MLflow, SageMaker, Databricks ML, Azure ML StudioData: Databricks, Snowflake, S3, Delta LakeVisualization: PowerBI, Tableau, Streamlit, DashExperimentation: Evidently AI, CausalML, upliftMLVersioning: Git, DVC, MLflow TrackingPreferred (Nice-to-Have)Experience with Telecom Offer Recharge Modeling or Dynamic Pricing Systems.Knowledge of Pricefx PriceAI, Adobe Target Recommendations, or Reinforcement Learning frameworks.Understanding of Elasticity Curves, Customer Lifetime Value (CLV), and Offer Fatigue Modeling.Experience integrating ML outputs into business decision engines or rule systems. Build and deploy models for:Price Elasticity / Conversion PredictionChurn Propensity / Retention UpliftSegment Discovery Similarity (Clustering, KNN)Offer Recommendation / Ranking (Scoring Models)Design A/B testing and uplift modeling to evaluate campaign performance.Develop simulation engines for pricing what-if analysis and scenario testing.Create automated pipelines for model training, scoring, and retraining.Work closely with Data Engineers to ensure feature store alignment.Collaborate with the Business Decisioning team to translate insights into rules and thresholds.Implement feedback loops using real-time events (purchase, rejection, expiry) to improve models. Build and deploy models for: Price Elasticity / Conversion PredictionChurn Propensity / Retention UpliftSegment Discovery Similarity (Clustering, KNN)Offer Recommendation / Ranking (Scoring Models) Price Elasticity / Conversion Prediction