Senior Specialist - Data Sciences

Results-driven Data Scientist with hands-on expertise in time series forecasting, statistical analysis, data analysis, data wrangling, and predictive modeling using Python and SQL. Strong understanding of forecasting techniques including trend, seasonality, and model evaluation to enhance forecast accuracy. Skilled at converting complex datasets and forecasting outputs into actionable business recommendations.

Key Skill Areas

Thorough understanding of time series concepts such as trend, seasonality, cyclic patterns, lag features, rolling statistics, anomalies, and historical patterns. Hands-on experience with forecasting models including Prophet, ARIMA, SARIMA, MSTL, ARIMA/TIMA TBATS, Exponential Smoothing, HoltWinters, and Theta models.

Proficiency in descriptive and inferential statistics, probability distributions, correlation and covariance, regression analysis, hypothesis testing, confidence intervals, variance analysis, statistical significance, outlier detection, and trend analysis.

Forecast Model Evaluation & Optimization: Evaluate forecast accuracy using MAPE and percentage-error-based metrics, perform time-based validation, compare models, conduct hyperparameter tuning, analyze overfitting and underfitting, identify forecast bias, and improve overall forecasting performance.

Exploratory Data Analysis (EDA): Perform trend and correlation analysis, conduct historical pattern analysis, manage missing values, handle outliers, create lag/rolling features, develop holiday and event features, and build external regressors/covariates for daily, weekly, and monthly data analysis.

Soft Skills

• Strong problem-solving ability and high attention to detail; capable of stakeholder communication and translating business problems into data solutions

Must Have

• Developed, maintained, and evaluated time series forecasting models to predict business performance using historical actuals, trends, seasonality, and statistical patterns
• Applied forecasting techniques such as Prophet, ARIMA, SARIMA, MSTL, ARIMA/TIMA TBATS and other statistical forecasting approaches to address business forecasting problems
• Performed forecast accuracy analysis, model comparison, and hyperparameter tuning to improve forecasting performance
• Conducted in-depth analysis of historical data to identify trends, seasonality, anomalies, forecast deviations, and key business drivers
• Performed forecast vs actual analysis and root-cause analysis to understand significant deviations and recommend model or data improvements
• Used Python and SQL for data preparation and transformation/feature development
• Incorporated holidays, events, and business covariates to enhance forecasting models
• Collaborated with stakeholders to gather forecasting requirements, translate business questions into analytical approaches, and communicate model results and forecast insights clearly

Good To Have

• Familiarity with prompt engineering, Copilot, OpenAI APIs, and basic usage of GPT/LLM models for analytical workflow automation
• Awareness of LLM limitations, responsible AI practices, and applications of GenAI in analytics
• Familiarity with advanced forecasting/ML approaches such as XGBoost, Neural Prophet, NBEATS, or ensemble forecasting
• Experience with Microsoft Fabric, Fabric Notebooks, Lakehouse, and Pipelines for running and operationalizing data science/forecasting workloads
• Willing to work in shifts

Mandatory Skills : Python - Data Science

Mandatory skill :Python, SQL, Time Series Forecasting, Statistical Analysis, Data Science