Senior Data Scientist – Load & Renewable Generation Forecasting

Position Overview

We are seeking a Senior Data Scientist to develop, optimize, and productionize forecasting solutions for wind generation, solar generation, and electric load.

The primary focus of this role is wind generation forecasting, with additional responsibilities supporting solar and load forecasting initiatives. This is a hands-on position combining data science, machine learning, time-series forecasting, meteorological analytics, renewable energy modeling, data integration, and production support.

The successful candidate will independently manage the forecasting lifecycle, including data exploration, feature engineering, model development, backtesting, validation, deployment, monitoring, and continuous optimization.

A key responsibility is evaluating and optimizing weather data sources and Numerical Weather Prediction (NWP) providers. The candidate will assess provider accuracy, geographic coverage, forecast horizons, available variables, latency, reliability, and redundancy to identify the most effective weather inputs for different forecasting applications. The role also involves developing provider-blending strategies, improving forecast resilience, and reducing unnecessary data-source overlap without compromising accuracy.

The ideal candidate combines advanced statistical and machine-learning expertise with a practical understanding of meteorological conditions, renewable generation systems, and the operational factors influencing electric demand and power generation.

Key Responsibilities

1. Forecasting and Machine Learning

• Develop, enhance, and productionize wind, solar, and load forecasting models.

• Build forecasting solutions using weather forecasts, historical generation and load, SCADA telemetry, operational data, statistical baselines, physical relationships, and machine-learning techniques.

• Apply regression, time-series analysis, gradient boosting, ensemble methods, and probabilistic forecasting.

• Perform feature engineering, model selection, hyperparameter optimization, validation, and performance benchmarking.

• Develop backtesting frameworks to evaluate forecast accuracy across locations, forecast horizons, seasons, and operating conditions.

• Investigate forecast errors and distinguish between weather-related inaccuracies, model limitations, data-quality issues, and operational impacts.

• Modernize existing forecasting models and simplify model architectures where practical.

2. Weather Provider Evaluation and Optimization

• Evaluate multiple commercial and public weather and NWP providers based on accuracy, reliability, latency, geographic coverage, forecast horizons, and available variables.

• Compare provider performance by site, weather variable, forecast horizon, season, and weather regime.

• Identify provider-specific strengths and determine which weather inputs deliver the greatest value for each forecasting application.

• Assess data-source overlap and recommend an efficient, reliable set of weather providers.

• Develop and evaluate weather-provider blending, ensemble weighting, provider selection, bias correction, feature selection, fallback, and substitution strategies.

• Validate weather data alignment across issue times, valid times, time zones, spatial resolution, units, and forecast horizons.

• Monitor weather-data quality and identify provider outages, anomalies, or performance degradation that could affect forecasts.

3. Renewable Energy and Load Forecasting

Wind generation forecasting

• Analyze wind speed and direction, hub-height adjustments, vertical wind shear, temperature, pressure, humidity, and air density.

• Apply turbine power curves, cut-in/rated/cut-out behavior, wake effects, and site-specific characteristics.

• Analyze SCADA telemetry, generation output, curtailment, derates, outages, maintenance, and equipment availability.

Solar generation forecasting

• Analyze solar irradiance, cloud conditions, ambient and module temperature, and other weather-related generation drivers.

• Account for inverter performance, equipment availability, tracker positioning, stowing, and plant-controller impacts.

• Consider curtailment, outages, maintenance, soiling, degradation, and available plant capacity.

• Integrate solar plant telemetry and operational data into forecasting models.

Electric load forecasting

• Model temperature, humidity, heat index, and other weather-related demand drivers.

• Analyze calendar effects, seasonal trends, lagged demand, and historical weather-demand relationships.

• Account for extreme-weather conditions, regional differences, customer segments, and peak-demand patterns.

• Distinguish weather-driven variability from operational constraints when training, validating, and evaluating forecasting models.

4. Model Performance and Production Reliability

• Monitor forecast performance using MAE, RMSE, normalized error, bias, forecast skill, and appropriate probabilistic metrics.

• Benchmark forecasts against persistence models, existing solutions, alternative weather providers, and third-party forecasts.

• Develop monitoring and diagnostic tools to identify forecast degradation, model drift, data-quality issues, provider anomalies, and operational impacts.

• Implement logging, alerts, error handling, and recovery procedures for production forecasting workflows.

• Design resilient forecasting processes that can continue operating when individual weather providers or data feeds become unavailable.

• Conduct root-cause analysis of significant forecast misses and implement corrective actions.

5. Data Integration and Production Deployment

• Develop data connectors and pipelines for weather, SCADA telemetry, historical generation, load, and operational data.

• Support scalable batch and near-real-time forecasting workflows.

• Maintain model configurations, versioning, dependencies, testing, and technical documentation.

• Collaborate with data engineers and DevOps teams to deploy, monitor, and maintain forecasting applications.

• Develop dashboards and analytical tools using Streamlit, Plotly, or comparable technologies to communicate forecast performance, model behavior, and operational impacts.

6. Collaboration and Continuous Improvement

• Collaborate with data scientists, data engineers, DevOps, operations, trading, and business stakeholders.

• Communicate forecasting results, model limitations, performance trends, and technical recommendations clearly.

• Support knowledge sharing, technical documentation, and mentoring of less-experienced data scientists.

• Identify opportunities to improve forecasting accuracy, operational reliability, and the efficiency of data and model workflows.

Required Qualifications

• Advanced proficiency in Python and data-science libraries such as pandas, NumPy, scikit-learn, statsmodels, LightGBM, and XGBoost, or equivalent tools.

• Demonstrated experience developing, validating, and deploying forecasting, time-series, or predictive machine-learning models.

• Strong understanding of feature engineering, regression, time-series methods, gradient boosting, ensemble forecasting, and model evaluation.

• Experience with backtesting, model benchmarking, and forecast performance metrics.

• Experience integrating and analyzing weather APIs, NWP forecasts, databases, SCADA/telemetry, or external data feeds.

• Ability to evaluate weather-provider performance and quantify its impact on forecast accuracy.

• Strong SQL and data-analysis skills.

• Experience moving models from research and development into reliable production workflows.

• Working knowledge of AWS services such as EC2, S3, Lambda, ECS, Step Functions, CloudWatch, and SageMaker, or comparable cloud platforms.

• Experience with Git, version control, software testing, and reproducible development practices.

• Strong analytical, troubleshooting, communication, and documentation skills.

• Ability to work independently and collaborate across technical and business teams.

Preferred Qualifications

• Experience with utility-scale wind forecasting or renewable generation modeling.

• Experience working with multiple commercial or public weather/NWP providers.

• Direct experience with wind turbines, solar plants, or SCADA telemetry.

• Familiarity with turbine power curves and weather-to-power conversion methodologies.

• Experience benchmarking internal forecasts against commercial forecasting vendors.

• Familiarity with energy markets, utility operations, generation scheduling, trading, or ISO/RTO environments.

• Experience developing intra-hour, day-ahead, and multi-day forecasts.

• Experience with probabilistic forecasting, quantile regression, and forecast uncertainty.

• Experience with production model monitoring, drift detection, and automated alerting.

• Experience developing analytical dashboards using Streamlit, Plotly, or similar tools.

• Experience mentoring junior data scientists.

Key Success Factors

• Improved wind generation forecast accuracy and reliability.

• Effective use of weather data and optimization of provider selection.

• Reliable production forecasting models across wind, solar, and load.

• Strong model monitoring, explainability, and root-cause analysis.

• Robust data integration and resilient forecasting workflows.

• Clear communication of forecast performance and actionable recommendations.