Description
Description
SAIC has an opportunity for a Research Scientist to provide technical support services in applying Machine Learning to Navy Numerical Weather Prediction (NWP) development, and testing. Work shall be performed in accordance with all applicable DSRC usage policies, security requirements, and Government-furnished technical direction.
This position is a remote position.
Task 1: High-Resolution Machine Learning Weather Model Testing and NWP Reanalysis: Evaluate emerging high-resolution AI weather prediction systems (such as Atmo.ai) for Navy-relevant maritime environments, and develop a high-resolution, multi-year COAMPS-based regional weather reanalysis dataset over the Western Pacific to support model evaluation and scientific study.
Task 2: Navy Machine Learning Global Weather Model (Validation and Transition Effort): Fine-tune and transition a GraphCast-based global machine learning weather model into FNMOC operations, evaluate additional emerging ML weather models, and coordinate the development of a lightweight regional ML model to improve maritime decision-making and forecasting.
Task 3: Multiscale Predictability of Atmospheric Rivers and Air-Sea Interaction: Investigate the fundamental physical processes, air-sea fluxes, and boundary layer influences that affect the multiscale predictability of atmospheric rivers using advanced Navy prediction systems, machine learning models, and targeted field observations (e.g., dropsondes and buoys) to improve high-impact weather forecasting operational suitability of new observation sources.
Qualifications
EDUCATION AND EXPERIENCE:
- Bachelors and two (2) years or more experience; Masters and 0 years related experience; PhD
- PhD in Atmospheric Sciences or a related scientific or engineering field of study
Shall have at a minimum of 3 years of demonstrated research experience in:
- Applying machine learning models to probabilistic weather forecasting
- Evaluating forecast skill and systematic biases in NWP reforecasts
- Improving decision-making under Identified regime-dependent uncertainty
- Evaluating AI climate emulators to reproduce and forecast stratosphere-troposphere coupling
- Programming proficiency in Python, R, and SQL
- Must have experience with a combination of AWS, Git/GitHub, High-Performance Computing (HPC), GPU-accelerated computing, ERA5/ERA-I, ECMWF and UFS reforecast, AI-based climate/weather emulators (GraphCast, ACE2).
Desired Experience:
- Experience publishing peer-reviewed scientific literature, software documentation, guidebooks, and/or handbooks.
- High Performance Computing (HPC) experience.
CLEARANCE REQUIREMENT:
- Candidate required a completed T3 Security Investigation for full access to government IT systems
Apply on company website