Field Energy

Senior Data Scientist (Forecasting)

Join Field Energy as they are looking for a Senior Data Scientist (Forecasting)

Job Description - Senior Data Scientist: Forecasting - Field


We are looking for a Senior Data Scientist to build robust forecasting models. This is a highly specialised role with a focus on R&D to keep us up to date with state-of-the-art methodologies. We can provide support in deploying the models, tracking their performance and crawling data. The models you will come up with will define our deployment strategy (not the other way around). We want someone to focus on the science rather than the engineering side of Data Science.

This role is critical to Field’s mission. We are building our battery optimisation platform from the ground up, so this is a very exciting time to join the crew. You will primarily be working with our forecasting data scientists (Beth and Andreas), as well as our data engineers and optimisation data scientists. Our forecasts form inputs into our optimiser, so there will be a lot of interaction with that team, and forecasting output requirements will be dependent on our optimisation methodology.

Below is an overview of the main responsibilities. However, we will grow rapidly as will your responsibilities, so this is by no means an exhaustive list. There will always be room to grow and learn.


  • Forecasting of cleared energy volume and price in different markets
  • This involves conventional time series forecasting, where the target variable follows seasonal patterns and is driven by market fundamentals
  • Confidence intervals: we are interested in more than point estimates. Confidence intervals/ probability distributions are critical inputs into our optimisation engine
  • Forecasting newer markets with fewer actors involved
  • Some markets are limited to specific asset operators, so the bidding strategy of these actors becomes more important than market fundamentals
  • Pay-as-bid forecasting
  • Quantile regression: for some markets we have a range of prices for each trading period as opposed to a single cleared price.
  • Propensity modelling: predicting the price of these markets is not enough, as depending on the location of our assets we have different acceptance probabilities due to the way National Grid operators balance supply and demand

A bit about you:

  • Enthusiasm and curiosity about the energy storage industry
  • Humble and hands-on team player, necessary in a small company environment
  • Ability to work well with others at all levels, including senior management
  • Ability to work to tight timescales and deadlines
  • Ability to multitask, manage time effectively, and plan across multiple initiatives
  • Great communicator that enjoys both upskilling team members and learning from them

Experience and skills we look for:


  • Have published in the time series domain or are comfortable reading academic papers and turning those into code
  • Proven track record of building robust forecasting models with measurable business impact
  • Up to date with some recent forecasting methodologies (attention based NN, LSTM, CNN)
  • Experience with probabilistic forecasting
  • Familiar with version control

Nice to have:

  • Experience with causal inference, especially in the time series domain
  • Have built robust backtesting and experimentation frameworks


We code in Python and use the following tools:

  • DS project framework: kedro
  • Data lake: Athena, s3
  • Orchestrator: Prefect
  • Infrastructure: Terraform, AWS
  • CICD: Docker, Github Actions

This is by no means an exhaustive list. We’re also hoping you will make some suggestions ;) .