Senior Data Scientist, PV Performance & Diagnostics

Quick Facts

• Role: Senior Data Scientist
• Focus: Physics-based performance modeling, losses/underperformance reporting, and expected-energy forecasting for solar PV and battery energy storage within an asset management platform

Description

Stem is seeking a Senior Data Scientist with hands-on experience in solar PV performance modeling to advance performance analytics for solar PV and battery energy storage in PowerTrack. You will design and standardize physics-based performance models, own underperformance and loss reporting accuracy, and develop time-series forecasting for expected-energy prediction. The role expects a strong foundation in statistics and machine learning with room to grow into more advanced ML-based techniques.

Responsibilities

• Lead design, implementation, and continuous improvement of physics-based performance models for solar PV and battery energy storage, including data configuration and validation across heterogeneous site layouts
• Own accuracy and defensibility of performance and losses/underperformance reporting (downtime, curtailment, soiling, and other loss categories) used for customer reporting and decision-making
• Develop and contribute to time-series forecasting models for expected-energy prediction to inform downstream merchant operations and storage/hybrid use cases
• Run structured experiments and validate model accuracy against known or proxy ground truth; produce technical documentation, validation reports, and benchmarks for technical and non-technical stakeholders
• Partner with software engineers and data engineers to integrate performance models and configuration tooling into production systems that are scalable, testable, and maintainable
• Collaborate with ML fault detection efforts; connect detected anomalies to underperformance detection and loss categorization and align on shared data and system interfaces
• Exercise independent judgment on prioritization and technical approach, escalate to subject matter experts when needed, and lead peer reviews of data science artifacts and code

Requirements

• 5+ years in data science, analytics, PV/solar performance engineering, or related quantitative field (3+ years with a PhD)
• Bachelor’s degree in Electrical Engineering, Renewable/Solar Energy, Physics, Data Science, Statistics, Computer Science, Mathematics, or related field (or equivalent experience)
• Hands-on experience with solar photovoltaic performance modeling using tools like pvlib or similar irradiance/PV simulation libraries
• Strong statistics and machine learning foundation: regression, classification, time-series analysis, hypothesis testing, and model evaluation
• Proficiency in Python with pandas, NumPy, scikit-learn
• Advanced SQL skills with large-scale time-series datasets
• Experience building, validating, and benchmarking production-grade models or statistical algorithms
• Experience writing well-documented, reproducible, and maintainable scientific code with Git/version control and modern AI coding assistants
• Strong communication and presentation skills to translate complex technical findings for diverse audiences

Benefits

• Competitive compensation package with eligibility for bonus or commission
• Full health benefits on the first day of employment (medical plan options, dental plans, FSA/HSA with employer contribution, employer paid vision/LTD/STD/Life; voluntary coverage)
• 401k (pre- or post-tax) on first day of employment
• 12 paid calendar holidays per year
• Flexible time-off