Boulder, Colorado, United States
I value meaningful connections and a network based on shared interests and mutual professional goals, rather than just numbers. If we haven’t met yet, please include a short note with your interests or how we might collaborate. ------------------------------------------------------------------------------------------------------------------------------> As Senior Vice President of Engineering at Xoople, I bring over two decades of experience advancing geospatial technologies, driving innovation at the intersection of satellite systems, artificial intelligence, and computer vision to redefine how we observe and understand our planet. I am dedicated to advancing the geospatial industry through innovations in AI, signal processing, super-resolution, data fusion, hyperspectral imaging, and synthetic aperture radar analysis. I am committed to the responsible use of AI and actively support sustainability and climate initiatives. I am a proud advocate of the SpaceNet program, which advances geospatial AI for societal benefit. My contributions to the field include over 100 publications, spanning patents, journal articles, book chapters, and peer-reviewed conference papers (h-index = 28). Beyond my role at Xoople, I serve as Vice President of Technical Activities for the IEEE Geoscience and Remote Sensing Society (GRSS), leading a global network of over 150 volunteers. My previous leadership roles within IEEE GRSS include Director of Corporate Relations, Editor-in-Chief of the GRSS eNewsletter, and Chair of the Data Fusion Technical Committee. I believe in the power of collaboration and mentorship to foster innovation, continuous learning, and cross-functional excellence. Through my work at Xoople and my service to the IEEE community, I am helping shape the future of satellite imagery and empowering the next generation of geospatial leaders.
Powering the next generation of Earth Intelligence and EarthAI
• Developed the HD product, industry-first to market 15 cm satellite imagery - HD imagery enables faster and more accurate data extraction from images collected by the Maxar satellite constellation, improving the clarity needed for advanced machine learning and analytics at global scale - machine learning models trained on HD images outperform models trained on native images by detecting features of interest with greater accuracy across real-world conditions. HD models demonstrated more than 5 to 10% improvement in F1 score over the native-resolution models across repeated tests and applications - in 2023, Maxar announced the first 30 cm, global basemap using satellite HD images thanks to innovations in tasking and processing. A globally consistent, high-resolution imagery basemap is a critical component for most mapping methodologies, feature extraction techniques, virtual simulations and as the visualization layer in mobile mapping applications. - in 2024, Maxar announced the first 15 cm high-definition satellite imagery basemap that covers areas of change and population around the world. The clarity of Vivid Advanced 15 cm HD imagery enables digitization of finer map details like road markings and utility lines while also providing realistic and valuable context for navigation and location-based services. Read about HD from: https://spacenews.com/maxar-upgrades-satellite-imagery-basemaps/ https://geoawesome.com/how-15-centimeter-satellite-imagery-is-changing-the-mapping-game/ • Developed deep learning prototypes to validate the quality improvement of object detection applications when using HD products, focusing on subjects beneficial to global initiatives. Particular attention was given to features valuable in supporting the United Nations Sustainable Development Goals (SDGs). • Design of algorithms, prototypes, and HPC software implementations aimed at improving Maxar pansharpened products • Supported the production of the Legion first light images
• Worked with product managers, scientists, and engineers to execute on customer projects • Provided technical expertise to project statements, proposals, and other reports • Supported customer technical training and meetings
• Developed deep learning proof of concept prototypes for object detection and classification, image denoising, and inpainting • Provided technical guidance for the position of the spectral bands of the upcoming Legion constellation • Technical lead and Product Owner (Agile) of a scrum R&D team of scientists, engineers and developers responsible for the surface reflectance product - The AComp surface reflectance product not only delivers enhanced quality imagery from the DigitalGlobe archive, but it also means more satellite capacity is available for tasking customers, especially in high demand areas of the world
• Technical lead of a team of scientists, engineers, and developers working on the absolute spectral accuracy of image products: - design of algorithms, prototypes, and HPC software implementations for the retrieval of surface reflectance (AComp) - verification, quality assessment, and A/B testing of the surface reflectance product - radiometric calibration and validation of the DigitalGlobe fleet • Responsible for generating weekly, quarterly, and ad-hoc reports for management review. • Scientific Review Board: - Corporate content review of highly technical, brand-building documents, including industry thought-leadership and educational articles, sales and marketing advertorials, and market vertical-specific advertising, presentation decks, social media posts, and industry blogs.
• Developed AComp, a fully automated atmospheric compensation technique for very high spatial resolution panchromatic, visible/near-infrared, and short-wave infrared data (patents US-9,396,528 and US-9,990,705 awarded) • Designed algorithms and software prototypes for cloud and haze detection • Conducted Cal/Val activities using in-situ measurements of ground targets and calibrated tarps over various geographical regions with different atmospheric conditions • Worked across teams within and outside of the R&D department to support the production, enhancement, and analysis of very high spatial resolution imagery: - machine learning systems to predict image quality, analysis of WorldView-2 compression rate, and charge transfer efficiency - pansharpening (patent US8761506 awarded), image superresolution, and boosting - image classification and change detection, analysis of large image time-series, feature extraction (cars, ships, illicit crops, and camouflage detection), analysis of multi-angular data, analysis of natural disasters events (oil spills, earthquakes, floods, and hurricanes), bathymetry and benthic habitat mapping
• Developed novel classification and change detection techniques for urban applications using very high spatial resolution optical and synthetic aperture radar imagery, with special emphasis on machine learning and data fusion • Developed colorization techniques for DigitalGlobe panchromatic WorldView-1 satellite