Senior ML Engineer / Data Scientist

• Industry: Energy System

• Location: preferably Poland

• Workload: full-time

• Model of working: 100% Remote

• Contract: B2B, initially for 6 months + extensions

Summary: Main focus here is to develop and refine time series forecasting models to optimise energy asset performance within the Client’s optimisation platform, leveraging Machine Learning and Data Science techniques.

Responsibilities:

• Develop and enhance time series forecasting models across multiple domains, including load, generation, and consumption.

• Conduct exploratory data analysis (EDA) to understand data behaviour and quality, assessing trends, seasonality, autocorrelation, and identifying and addressing missing values or outliers.

• Apply and compare both statistical and machine learning approaches to build robust, scalable forecasting pipelines.

• Evaluate, monitor, and benchmark model performance using appropriate metrics to ensure model robustness over time.

• Collaborate closely with optimisation and engineering teams to integrate forecasts into production decision systems and measure their real-world impact on asset performance.

  
  

Must Haves:

• Strong understanding of time series forecasting principles, including univariate and multivariate modeling, and effective use of exogenous covariates.

• Solid knowledge of exploratory data analysis (EDA) techniques for time series.

• Ability to frame forecasting as a machine learning problem, including both regression-based and window-based prediction approaches, and to evaluate results rigorously.

• Familiarity with modern time series forecasting frameworks and libraries, along with sound model evaluation and selection.

• Proficiency in Python and experience building cloud-based ML pipelines (preferably on GCP, including Vertex AI).

  
  

Nice to Haves:

• Experience with data visualisation tools to communicate insights effectively.

• Knowledge of optimisation algorithms relevant in energy markets.

• Familiarity with additional programming languages relevant to machine learning applications.

  
  

Other Details:

• Team Structure: Small team with a bunch of models already running, with no business tasks/contact with stakeholders.

• Project Context: Involves real-time decision-making systems for energy asset management.