GalaxEye Space: Sync Fused Satellite Imagery
GalaxEye Space is pioneering the world’s first SAR+MSI satellite, capable of capturing the same Area of Interest with both sensors simultaneously and fusing them into a single “sync-fused” image. This innovation overcomes limitations like cloud cover, where SAR data can replace obstructed optical imagery, delivering more reliable datasets for analysis.
As the Product Manager, I translated these needs into an actionable roadmap—establishing tile sizing and metadata standards, scoping the MVP into modular capabilities, coordinating cross-functional teams, and running tasking simulations—laying the foundation for consistent, scalable imagery delivery.
The goal was to replace fragmented processes with a scalable system by standardizing formats, defining product structures, and enabling efficient discovery, processing, and delivery. This involved resolving technical inconsistencies, introducing a robust tasking framework, and ensuring adaptability for future satellites and partners.
Problem Statement
-
GalaxEye Space required a unified platform to manage and deliver imagery from multiple sensors, including its pioneering SAR+MSI sync-fused data.
-
Existing workflows were fragmented, with varying formats, unclear product definitions, and inconsistent delivery mechanisms.
-
These gaps caused delays in integration, onboarding, and scaling for future satellites and partners.
-
The core challenge was to design a scalable platform capable of handling the full pipeline—from capture to delivery—while remaining adaptable to future mission needs.
M A J O R I N S I G H T S
Data Distribution platform
A strong satellite data platform starts with a solid foundation for storing and finding information.
This means using formats that are easy to access and search, while keeping imagery and its details separate for faster retrieval. Storing imagery in cloud-friendly formats linked to a central catalog allows users to work directly in the cloud, avoiding repeated large downloads and enabling quicker analysis.
We applied a modular, step-by-step approach: first managing how data enters the system (ingestion), then organizing storage and search, adding processing pipelines, and finally delivering the data to users. This method sped up development, ensured each component was robust, and created a scalable structure for future growth.
Product Analysis
Clear and consistent product definitions are essential for smooth integration. By fixing the size of each imagery tile, standardizing naming rules, and using consistent metadata, most of the common issues between teams and customers can be eliminated. This approach ensures that every image product is predictable, easy to reference, and simple to integrate into different systems.
It also helps to think of satellite imagery as two connected layers. The first is the catalog or discovery layer, where users can browse and preview what’s available. The second is the actual data layer, containing ready-to-use, high-quality imagery with metadata. Together, these layers make the platform both easy to explore and reliable to use.
Satellite tasking research & development
While every satellite tasking API has its own style, they often follow the same basic structure. Leading providers like Capella, Maxar, Umbra, and Airbus all use similar building blocks: a way to define the target area, set collection priorities, choose time windows for imaging, and check the status of a request.
Adopting this common structure for an internal API makes future integrations much smoother and reduces the need for major adjustments later.
Equally important is testing these processes through simulations. Running weekly and quarterly tasking simulations reveals real-world constraints—such as orbit timing, revisit rates, and weather impacts—early in the planning stage. This helps mission and product teams set realistic expectations and deliver on commitments.
Challenges in Satellite Imagery Delivery
Capturing satellite imagery is only part of the challenge—getting it to the end user quickly is equally critical. Delivery speed often depends on ground-station availability and the network that transfers data from the ground to the cloud. When these links are slow or congested, delays can occur. To address this, new methods like store-and-forward data handling and high-speed data transfer networks are being explored.
Once in the cloud, distribution brings its own hurdles. Costs can rise quickly due to data transfer fees, multi-region storage, and compliance with regional data regulations. A practical solution is to let customers process data where it is stored or provide smaller, preview-ready tiles instead of moving entire datasets.
