Seasonal Research Geophysicist Jobs (NOW HIRING)

SAIC is seeking an experienced Scientific AI programmer for a position at NOAA's Geophysical Fluid Dynamics Laboratory (GFDL) that will focus on creating a complex earth system model emulator. Through the work of this position, SAIC will develop a seasonal to subseason (S2S) AI weather forecast model.  The Scientific AI programmer will evaluate an existing emulator and develop a prototype for the S2S timescale.  You will work collaboratively with federal, contractor, university, and private sector scientists and developers and with the scientific evaluation team to assess the readiness of the AI forecast model for operational use.

This position requires an ability to obtain and maintain a Public Trust background investigation. This position is located in Princeton, NJ. 

Responsibilities include, but are not limited to: 

• Evaluate the existing seasonal to decadal AI emulator prototype to determine its utility for S2S timescales
• Build an S2S AI emulator using SPEAR hindcasts, reanalysis, and other earth-system data for S2S forecasts 
• Determine the operational readiness of the S2S AI emulator 
• Coordinate with federal, contractor, university, and private sector scientists to align variables and training frameworks with research objectives
• Deliver status updates through various communication channels such as team meetings and written reports

SAIC is a premier mission integrator focused on advancing the power of technology and innovation to serve and protect our world. Our robust portfolio of offerings across the defense, space, intelligence, and civilian markets includes secure high-end solutions in mission IT, enterprise IT, engineering services, and professional services. We integrate emerging technology, rapidly and securely, into mission critical operations that modernize and enable critical national imperatives.

Required Education: 

• Bachelors and five (5) years or more experience; Masters and three (3) years or more experience; PhD and 0 years related experience

Required Qualifications: 

• A Bachelor's degree in Computer Science, Information Systems, Engineering, Business or other related scientific or technical discipline.
• Two years of experience in developing and training AI models using large data sets
• Four years of specialized experience in determining information technology effects on the organizational structure and determining the ability that IT can support/meet organizational goals.
• With a Master's Degree (in the fields described in Min. Education above): two years experience
• With a PH.D. (in the fields described in Min. Education above).: no experience required
• With at least 8 years of specialized experience, a degree is not required.
• Proficient in Python Programming 
• Experience managing projects with Git
• Strong interpersonal skills to support collaborative team environments

Desirable Skills: 

• Basic knowledge of ocean, atmosphere and/or climate and weather processes or a related science.
• Experience with object storage and traditional disk environments
• Experience using NetCDF and Zarr datasets
• Familiarity with High-Performance Computing (HPC) environments and batch queuing systems like Slurm.
• Experience with modern AI-assisted coding workflows and/or MLOps tools to accelerate development cycles.

Background:

The Allen Institute for Artificial Intelligence (Ai2) and Multiscale Machine Learning In Coupled Earth System Modeling project (MLInES), in partnership with GFDL, are developing several AI emulators for long time-scale simulations.  These include an atmosphere model (ACE), an ocean model (Samudra), and a coupled model emulator (SamudrACE).  The Software Engineering for Novel Architectures (SENA) initiative funded GFDL in FY2026 to develop an AI model emulator at seasonal to decadal timescales using data from GFDL's SPEAR mode.  The SPEAR emulator project began in January 2026.

Program and Project Details: 

The proposed work is a continuation of the previous research of SamudraACE and the SPEAR emulator to apply these methods to the subseasonal to seasonal (S2S) timescales.  The work will be done primarily on NOAA systems. When ready, the S2S AI forecast will be transitioned into the NOAA EAGLE Project pipeline for operational use.