Signal Analysis Engineer III

Job Details

New ideas are all around us, but only a few will change the world. That’s our focus at JPL. We ask the biggest questions, then search the universe for answers—literally. We build upon ideas that have guided generations, then share our discoveries to inspire generations to come. Your mission—your opportunity—is to seek out the answers that bring us one step closer. If you’re driven to discover, create, and inspire something that lasts a lifetime and beyond, you’re ready for JPL.

Located in Pasadena, California, JPL has a campus-like environment situated on 177 acres in the foothills of the San Gabriel Mountains and offers a work environment unlike any other: we inspire passion, foster innovation, build collaboration, and reward excellence.

Responsibilities:

We are looking for a Signal Analysis Engineer III/Algorithm Developer to join the Multi-sensor Worldwide Ocean Winds (MWOW ) team. MWOW is a NASA Satellite Needs Working Group (SNWG) Cycle 4 solution, selected in response to one of the top five inter-agency needs identified in the 2022 survey, and it is funded through May 2030. MWOW combines ocean surface wind retrievals from every useful satellite scatterometer, radiometer, and synthetic aperture radar into a single coregistered, globally gridded product, delivered 6-hourly on both a 0.125-degree and a 1/64-degree global grid and archived for public distribution at NASA's Physical Oceanography Distributed Active Archive Center (PO.DAAC). Its users are the operational and research communities at NOAA and the Department of Energy, who need finer spatial resolution, better coverage near coasts and sea ice, and a consistent multi-decade record.

A Version 0.2 beta product drawing on seven satellites is available now for stakeholder evaluation. Version 1, scheduled for release in September 2027, expands the product to eight scatterometers (ASCAT-A/B/C, EOS-6, HY-2B/C, ScatSAT-1, and QuikSCAT), six radiometers (SMAP, GMI, AMSR2, and SSMI-F16/F17/F18) together with the historical TMI and SSM/I F-series, and three synthetic aperture radars (Sentinel-1, NISAR, and the SWOT Ka-band radar interferometer), and reprocesses the historical record back to 1999. Versions 2 and 3 follow in September 2028 and September 2029, with a wind stress and wind derivative product in March 2030. Reaching those deliveries requires new retrieval and correction algorithms, and a large share of them are machine learning algorithms.

The MWOW algorithms are described in the MWOW Algorithm Theoretical Basis Document (ATBD), which lays out a ten-step per-sensor processing chain and identifies where each algorithm still has to be built. Several of the hardest remaining problems are learning problems: estimating sea-ice concentration and correcting winds near the ice edge, retrieving wind direction from sensors that measure only speed, correcting rain contamination in Ku-band backscatter, harmonizing sensors with different resolutions and calibrations onto a common scale, and filling the spatial and temporal gaps between satellite overpasses. The coastal and near-sea-ice product plans additionally call for using machine learning to raise the effective resolution of scatterometer and radiometer retrievals by training against MWOW's own collocations with high-resolution SAR observations. The team has separately proposed a machine-learning augmentation to MWOW covering kilometer-scale super-resolution of the wind field, blending of SWOT wide-swath and conventional nadir altimeter sea surface height, and derivation of total ocean surface currents. That augmentation is not yet funded and is not the primary emphasis of this position, but we want someone who could take a leading role in it if it is.

Signal Analysis Engineer III will need to be comfortable at the intersection of ocean microwave remote sensing and modern machine learning, and who can turn a research algorithm into tested, documented, open-source software that runs in an operational production system.

The successful applicant will report administratively to the Group Supervisor of Group 363L and programmatically to the MWOW Task Manager.

Tasks may include but are not limited to:

Design, implement, test, and validate algorithms and ground processing software that produce the MWOW Level 3 ocean surface wind data products distributed to NOAA and Department of Energy stakeholders and to the public through PO.DAAC

Algorithms may include: per-sensor neural networks that estimate sea-ice concentration and correct winds near the ice edge; neural-network wind retrieval from Ka-band and L-band radar measurements; wind direction retrieval from speed-only SAR and passive microwave observations; coastal land-contamination correction of backscatter before retrieval; neural-network rain correction for Ku-band scatterometers; cross-sensor speed harmonization; per-retrieval uncertainty models built by triple collocation; and quality control and regridding of the merged multi-sensor product.

Apply machine learning to raise the effective resolution of the wind product near coasts and sea ice and over the open ocean, and to fill spatial and temporal gaps between satellite overpasses, using MWOW's collocations between coarser scatterometer and radiometer winds and high-resolution SAR observations as training data

Develop the collocation, simulation, and analysis tools needed to train and validate these algorithms, and characterize their performance against buoy networks, the ASCAT reference standard, and ERA-5 reanalysis

Deliver the work as tested, documented, open-source Python software, developing collaboratively on GitHub with feature branches, pull requests, and code review in the existing MWOW repositories

Work collaboratively with an international team, including partners at KNMI, Remote Sensing Systems, PO.DAAC, and the NASA Ocean Vector Winds Science Team, to develop, test, and deliver complex algorithms and software systems

Understand and apply principles, theories, concepts, and techniques in microwave remote sensing, radar and radiometer signal processing, and machine learning

Document algorithms in the MWOW ATBD, support project milestone reviews, and communicate results to agency stakeholders and to the scientific community

Qualifications:

Bachelor’s degree in Electrical Engineering, Geophysics, Physics, Applied Mathematics, Computer Science, or a related discipline with a minimum of 6 years of related work experience; Masters in a related discipline with a minimum of 4 years of related work experience; PhD in a related discipline with 2 years of related work experience.

Understanding of radar and radiometer remote sensing as applied to the ocean, such as scatterometry, synthetic aperture radar, radar altimetry, or passive microwave radiometry, including how geophysical parameters are retrieved from backscatter or brightness temperature

Basic understanding of ocean surface winds and their relationship to atmospheric and oceanic process

Demonstrated ability to develop, train, and validate machine learning models for geophysical retrieval, regression, or image-to-image problems using a standard framework such as TensorFlow or PyTorch

Strong Python development skills, including the scientific stack (numpy, scipy, xarray, netCDF4), and collaborative software development using Git and GitHub

Data analysis skills, including experience working with large swath and gridded geophysical datasets

Strong communication skills.

Ability to work well in a team environment.

Preferred Skills:

Experience with deep learning architectures for super-resolution, gap filling, and data fusion, such as neural operators, score-based diffusion models, state-space and transformer architectures, or with physics-constrained learned solvers such as 4DVarNet and physics-informed neural operators

Hands-on experience retrieving ocean surface winds from scatterometer, SAR, or passive microwave o