Responsibilities
- Design, develop, and deploy high‑quality time series forecasts for energy market prices, ancillary services, energy demand, and renewable generation (e.g., solar PV and wind).
- Apply a broad range of forecasting methodologies, including classical statistical models, machine learning, and deep learning approaches, selecting methods appropriate to data regimes and business constraints.
- Lead feature engineering efforts incorporating calendar effects, weather signals, exogenous drivers, and regime changes.
- Establish rigorous model evaluation, backtesting, and benchmarking frameworks to ensure accuracy, robustness, and stability over time.
- Architect, build, and maintain end‑to‑end MLOps pipelines, covering data validation, training, versioning, deployment, monitoring, and retraining.
- Ensure forecasting systems are scalable, observable, and reliable in production, with clear SLAs, alerting, and rollback strategies.
- Partner in the design and evolution of an internal forecasting platform that supports the full machine learning lifecycle and multi‑model production hosting.
- Implement best practices for model governance, reproducibility, experiment tracking, and lineage.
- Conduct applied research to identify new modeling techniques, architectures, and tooling that improve forecast accuracy, latency, and operational efficiency.
- Translate research ideas into production‑ready solutions, balancing innovation with maintainability.
- Influence technical roadmap decisions related to forecasting systems, data platforms, and MLOps standards.
- Work closely with engineering, product, and domain experts to ensure forecasting solutions deliver measurable business and operational impact.
- Incorporate energy system constraints and domain knowledge into models to ensure outputs are physically meaningful and actionable.
- Support production operations by troubleshooting issues, analyzing model degradation, and continuously improving system performance.
Basic qualifications
- Master’s or Ph.D. in statistics, machine learning, applied mathematics, computer science, or a related quantitative field.
- 5+ years of hands‑on experience in data science or machine learning, with significant exposure to time series forecasting in production.
- Strong proficiency in Python and experience writing production‑quality, maintainable code using modern software engineering practices.
- Deep theoretical and practical knowledge of time series methods, including statistical, regression‑based, and deep learning approaches.
- Demonstrated experience building and operating ML systems in production, including CI/CD for models, monitoring, and lifecycle management.
- Experience with cloud‑hosted platforms (preferably Azure / Fabric), containerization, and distributed compute.
- Proficiency with core data science and ML libraries such as pandas, numpy, statsmodels, sklearn, xgboost, lightgbm, pytorch, keras, and modern forecasting libraries (e.g., Nixtla).
- Strong problem‑solving skills, ownership mindset, and ability to operate effectively in ambiguous, real‑world environments.
- Travel may be required up to 10%, depending on business needs
Preferred qualifications
- Experience with energy systems, electricity markets, or infrastructure forecasting, including demand, pricing, or renewable generation.
- Familiarity with power systems concepts such as unit commitment, economic dispatch, or grid constraints.
- Prior experience designing or contributing to forecasting platforms or shared ML infrastructure.
- Exposure to large‑scale data pipelines, streaming or batch processing, and data quality frameworks.
- Experience collaborating across data science, software engineering, and operations teams in a production environment.
- This target salary range is for CA positions only and should not be interpreted as an offer of compensation.
Tags & focus areas
Used for matching and alerts on DevFound Fulltime Machine Learning Data Science Mlops Ai