Research Software Engineer

At Zeus AI, we're building an AI platform for Earth observations, supported by NASA, the Department of Energy (DOE), and the Department of Defense (DOD). Our interdisciplinary team of engineers and scientists is dedicated to a mission: to create a large-scale foundation model that will transform data assimilation, weather forecasting, and diverse scientific and commercial applications. Advised by industry-leading experts, our core objective is to enhance our understanding and management of the planet through research.

We are a remote-first company offering in-person work in Cambridge for team members located nearby.

Zeus AI integrates multi-modal, multi-resolution data to create a unified model of our planet. Our model predicts timely, complete, high-resolution global information. Our technology, developed at NASA Ames Research Center in Silicon Valley, uses state of the art computer vision to produce accurate, real-time and high-resolution weather variables. Our team has numerous publications and awards at the intersection of machine learning and Earth sciences with experience working in government agencies, academia, and private industry.

Zeus AI is building foundation models for global and regional scale Earth system modeling, ingesting observations to produce an accurate low latency representation of the planet through machine learning data assimilation. This problem is often framed as observation to observation forecasting. We are tackling this problem with a multi-modal and multi-resolution modeling framework using numerous observation types including satellites, stations, aircraft, drones, and marine vessels to power forecasts and digital twin models.

Data from this system must be ingested both historically and in near-real time while efficiently serving Earth Net outputs to users.

We are looking for a research software engineer to join our core science and engineering team. In this role, you'll be instrumental in building foundation models for data assimilation and forecasting. You will develop and manage the software infrastructure supporting machine learning operations including data and inference pipelines. This will include a combination of data engineering and machine learning operations for global and regional scale weather forecasts, optimizing system architecture and implementing best practices across the stack.

You will develop data pipelines to ingest remote sensing observations into Zeus’s data lake while optimizing data loading into machine learning pipelines. You will also manage serving data to users through cloud storage, APIs, and web visualizations.

• Design and implement robust pipelines using Xarray to ingest, process, and analyze multi-dimensional remote sensing datasets (e.g. lidar, radar, hyperspectral) from external public and private providers
• Manage machine learning inference infrastructure for weather forecasting and remote sensing data processing
• Support data infrastructure on a combination of cloud and high performance computing systems, utilizing distributed computing and serverless systems
• Design and maintain systems to monitor and report on pipeline health, data integrity, and computational performance
• Collaborate to serve Zeus AI data products to users through application programming interfaces and cloud storage systems
• Create data visualizations of our analysis and forecast predictions to communicate with technical and non-technical audiences

• Ph.D/M.S or B.S. with 2+ years of experience in computer science, remote sensing, physics, mathematics, or related quantitative field
• 2+ years of experience working with remote sensing datasets including infrared, hyperspectral, soundings, radar, and/or lidar observations
• Experience developing data visualizations of Earth science data
• Experience with Python, PyTorch/Tensorflow
• Expertise using the Pangeo stack, specifically Xarray/Zarr for complex earth science data
• Experience with high-performance computing systems using…