Statistical Data Scientist at Spire Global
Boulder, CO, US

Spire Global is seeking a Statistical Data Scientist to contribute to the Company’s effort of developing its portfolio of machine learning and statistical post-processing models for supporting our weather applications. This is an exciting opportunity for motivated scientists to improve the weather forecast through state-of-the-art predictive modeling, machine learning, statistical models, time series analysis, and numerical optimization techniques using the available surface observations, numerical weather predictions models outputs, large amount of Radio Occultation (GNSS-RO), Reflectometry (GNSS-R) and other satellite data.

As a successful candidate, you will join the Statistics and Machine Learning group under the Global Validation Model branch, working with top-level scientists at Spire and around the world. Collaborating with teams such as software engineering, modeling and data assimilation, you will be a part of implementing and evaluating machine learning and statistical weather models and making use of large datasets of surface and satellite observations.

Responsibilities will include the following tasks:

  • Propose and implement innovative predictive modeling, machine learning and Bayesian inference approaches to increase the forecast skills of weather fields forecasts and distributions.
  • Develop and implement advanced quality control methods.
  • Develop and implement advanced sampling methods to address insufficient data sets.
  • Develop and implement state of the art weather post-processing techniques.
  • Explore different standard and non-standard data sources including Spire GNSS-RO and GNSS-R measurements, visible and IR satellite images from the GOES Advanced Baseline Imager, feature selection and sensitivity techniques to identify predictors for weather data driven models.      
  • Working with meteorological data sets from various sources.
  • Working with the software engineering team to define most effective software solutions including transition to operations.
  • Presenting research findings at scientific conferences or workshops.


  • Applicants must have either a MS or PhD degree in Data Science, Computer Science, Applied Mathematics, or Atmospheric Science or equivalent working experience in machine learning algorithms and advanced statistical inference methods.
  • Working experience with linear/non-linear regression and classification methods and deep learning techniques including Generalized Linearized and Additive Models, Decision Trees, Clustering, Support Vector Machine, Neural Network, Multi-level models, Random Forests, Gaussian Processing, etc.
  • Working experience with time series analysis, recurrent and LSTM neural network.
  • Working experience with combining multiple models and ensembles for predictive modeling and uncertainty quantification (Mixture models for density estimations, Bayesian Modeling Averaging, Ensemble Model Output Statistics, etc.)
  • Working experience with statistical inference methods for uncertainty quantification such as Bayesian inference and Adaptive Metropolis algorithms.
  • Working experience with numerical optimization and automatic differentiation techniques.
  • Working knowledge of Python, Fortran, Matlab, Linux scripting and code management practices.
  • Prior experience working with meteorological or oceanographic datasets (GRIB and NetCDF formats) in distributed computing environments.
  • Experience with modern software engineering principles and best practices including DevOps environment.
  • Demonstration of enthusiasm and ability to work in a development team that never stops improving predicting modeling techniques, statistical and machine learning methods and their outcomes.