Senior Data Scientist (Machine Learning)

Senior Data Scientist (Machine Learning)

Location: Houston, TX

ABOUT FLUENCE 

Fluence, a Siemens and AES company, is a global market leader in energy storage products and services, and digital applications for renewables and storage. The company has more than 3.4 GW of energy storage deployed or contracted in 29 markets globally, and more than 4.5 GW of wind, solar, and storage assets optimized or contracted in Australia and California. Through our products, services and AI-enabled Fluence IQ platform, Fluence is helping customers around the world drive more resilient electric grids and have a more sustainable future. To learn more about Fluence, please visit: fluenceenergy.com 

ABOUT THE POSITION:

As a Senior Data Scientist – Machine Learning, you will provide technical expertise and collaborate with cross-functional teams to advance the forecasting capabilities within our market-leading optimization and bidding product, Mosaic, designed for renewable energy and battery storage assets in global electricity markets. You will identify impactful initiatives, address challenges proactively, and drive key projects to successful completion. In addition, you will contribute to mentoring and supporting the growth of team members within the Forecasting team.

Key responsibilities include:

•Experience: Bachelor’s degree with 5+ years, Master’s degree with 3+ years, or PhD with 2+ years of industry experience developing forecasting models, preferably in energy systems.

•Forecasting Model Development: Design, develop, deploy, and maintain advanced statistical and machine learning models for time series forecasting, with applications in demand, price, and renewable energy prediction. Incorporate probabilistic scenario generation to characterize uncertainty in forecasts.

•Advanced Forecasting Techniques: Leverage ensemble-based forecasting methods and develop self-correcting models of models that iteratively improve forecast accuracy by dynamically integrating feedback and error correction mechanisms.

•Uncertainty Characterization: Implement stochastic processing models to generate probabilistic forecast scenarios that characterize uncertainty in time series data.

•Multivariate Time-Series Modeling and Dependence Structure Analysis: Implement advanced stochastic modeling frameworks to capture the dependence structure of temporal and spatial correlations in time-series data across a sequence of markets, ensuring realistic scenario generation that reflects the complex relationships and dynamics within and between markets.

•Analyzing Electricity Market Dynamics: Conduct in-depth analyses of electricity market trends to decipher the evolution of market forces and identify shifting dynamics that influence price formation for energy and ancillary services. Leverage this understanding to refine forecasting models and strategic decision-making.

•End-to-End Workflows: Architect and manage workflows for feature engineering, model training, and inference using modern orchestration tools (e.g., Argo Workflows, AWS Step Functions).

•Model Deployment: Create, deploy, and optimize containerized solutions (e.g., Docker images) for scalable training and inference environments. Utilize Kubernetes and serverless platforms (e.g., AWS Lambda) for efficient model deployment.

•Data Engineering: Design robust ETL pipelines and define database schemas to streamline data ingestion, preprocessing, and storage, ensuring seamless integration into forecasting workflows.

•Infrastructure and Automation: Implement CI/CD pipelines for model and code deployment, ensuring repeatable, automated workflows for version control, testing, and production rollouts (e.g., CircleCI).

•Performance Monitoring and Optimization: Monitor deployed models and workflows to ensure reliability, scalability, and optimal performance. Continuously refine and improve systems to meet evolving requirements.

•Collaboration: Work closely with cross-functional teams, including data scientists, engineers, and product stakeholders, to ensure smooth integration of forecasting systems into broader applications.

What will our ideal candidate bring to Fluence? 

•Time Series Expertise: Strong understanding of time series forecasting techniques, including statistical models, machine learning algorithms, and deep learning approaches.

•Uncertainty Characterization: Expertise in implementing stochastic processing models to generate probabilistic forecast scenarios that characterize uncertainty in time series data.

•Advanced Forecasting Techniques: Familiarity with ensemble-based methods and self-correcting forecasting models that improve through iterative feedback.

•Programming and Software Practices: Advanced proficiency in Python for feature engineering, model development, training, and forecasting, with strong knowledge of modern software development practices, including version control (Git), testing frameworks, and agile methodologies.

•Agile/Lean Product Development: Experience working and delivering products or services in an agile/lean environment, demonstrating adaptability and efficient collaboration.

•Collaboration Skills: Demonstrated ability to collaborate with cross-functional teams and build strong working relationships across disciplines.

•Communication Skills: Excellent communication skills, with the ability to articulate technical concepts clearly and effectively to diverse stakeholders.

•Educational Background: An advanced degree (Master’s or PhD) in Computer Science, Operations Research, Electrical Engineering, Mathematics, Statistics, or a related field.

Preferred Qualifications:

Energy Industry Knowledge: Comprehensive knowledge of the energy industry, with a focus on deregulated electricity markets such as NEM, CAISO, ERCOT, MISO, PJM, and Japan.

•Cloud and Infrastructure: Proficiency with AWS services (e.g., SageMaker, Lambda, Step Functions), containerization tools like Docker, container orchestration using Kubernetes, and managing workflows for large-scale data and model pipelines.

•CI/CD Pipelines: Hands-on experience with continuous integration and delivery tools to automate and streamline model and software deployment.

•ETL and Data Management: Proven experience in designing ETL pipelines, creating efficient data schemas, and managing data flows at scale.

•Multivariate Time-Series Modeling: Experience developing stochastic frameworks that capture temporal and spatial correlations across markets, ensuring realistic and robust scenario generation.