- Design and implement statistical and machine learning models for time-series forecasting, anomaly detection, and asset health scoring across utility networks.
- Build and maintain end-to-end ML pipelines on Databricks from feature engineering and model training to validation, deployment, and monitoring in production.
- Apply classical statistical methods (GLMs, GAMs, mixed-effects models, Bayesian inference) alongside modern ML techniques (ensemble approaches, network analysis, neural networks) to solve grid operations problems.
- Develop predictive maintenance and degradation models for utility infrastructure using telemetry and SCADA data at scale.
- Translate ambiguous business problems into well-defined modeling problems with appropriate statistical frameworks - e.g., knowing when a LM/GLM is sufficient and when gradient boosting or deep learning is warranted.
- Implement model monitoring, drift detection, and automated retraining workflows to maintain model performance over time.
- Contribute to load forecasting, demand response optimization, and outage prediction systems.
- Ensure model interpretability and explainability for utility stakeholders and regulatory compliance.
- Contribute to internal knowledge-sharing on statistical best practices.
Requirements
- Bachelor’s degree or equivalent in Statistics, Applied Mathematics, Physics, Engineering, Data Science, or a related quantitative field.
- 3+ years of experience (with Bachelor’s), 2+ years of experience (with Masters), or 1+ years (with PhD) in applied statistical modeling and machine learning, with a track record of deployed production models.
- Strong programming skills in Python (PySpark, pandas, NumPy, scikit-learn, statsmodels, XGBoost), R (tidyverse, lme4, glmmTMB, glmnet, mgcv), and SQL for large-scale data analysis.
- Experience with time-series modeling (ARIMA, state-space models, LSTM, Darts, or similar) on high-volume meter data.
- Exposure to Databricks ML ecosystem (Feature Store, Experiment Track, Model Serving, Mosaic AI) and MLflow.
- Familiarity with distributed computing concepts - PySpark, Optuna/Ray, Spark SQL, partitioning strategies, and medallion architecture.
- Understanding of software engineering principles - version control (Git), testing, CI/CD for ML systems.
- Ability to communicate complex statistical/ML concepts to non-technical stakeholders.
Core Competencies
Demonstrates expertise in statistical modeling and machine learning for time-series forecasting and anomaly detection, with a strong focus on building and maintaining ML pipelines and ensuring model performance and interpretability. Proficient in translating business problems into statistical frameworks and communicating complex concepts to stakeholders.
Highest-signal resume keywords
- Statistical Modeling
- Machine Learning
- Python Programming
- Time-Series Modeling
- Databricks ML Ecosystem
ATS Optimization Keywords
Hard Skills
- Statistical Methods
- Machine Learning Techniques
- Feature Engineering
- Model Training
- Model Validation
- Model Deployment
- Model Monitoring
- Anomaly Detection
- Predictive Maintenance
- Data Analysis
Soft Skills
- Communication
- Problem-Solving
Industry Keywords
- Utility Networks
- Telemetry Data
- SCADA Data
- Regulatory Compliance
- Grid Operations
Tools & Technologies
- Databricks
- PySpark
- SQL
- MLflow
- Git