SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Planet designs, builds, and operates the largest constellation of imaging satellites in history, delivering empirical geospatial data via a cloud-based platform to commercial, environmental, and humanitarian sectors. The Analytics Team for Global Monitoring focuses on novel geospatial and time series analytics for Defense and Intelligence applications.
As a Senior Machine Learning Engineer, you will drive hands-on engineering and modeling for Defense and Intelligence use cases. You'll implement novel embeddings-based change detection and advanced computer vision techniques to extract insights from satellite imagery at continental and global scales. Key responsibilities include spearheading development of novel algorithms and ML models tailored for defense applications, optimizing model performance for high-throughput inference, innovating computer vision and time series techniques, collaborating with product managers and data scientists on algorithm design, integrating ML pipelines with software platforms, and establishing testing and monitoring frameworks for model reliability.
You bring 10+ years of relevant experience with 6+ years in machine learning. Required expertise includes data science, time series methods, computer vision, and embeddings. You're proficient in implementing and optimizing neural networks, wrangling large datasets with geospatial libraries, and using PyTorch/TensorFlow. You write clean, modular Python code following software best practices (Git, testing, CI/CD) and have experience deploying models via Docker/Kubernetes at scale. AWS or GCP experience is required. You hold a graduate degree in STEM or analytics or equivalent work experience, excellent communication skills, and ability to obtain US Security Clearance. Location in DC metro or ability to commute to Arlington 3x/week is required.
Standout candidates bring practical knowledge of remote sensing and satellite imagery, familiarity with coordinate reference systems and geospatial data formats (GeoTIFF, GeoJSON), hands-on experience building geospatial data products, and knowledge of model compression and GPU optimization techniques.
