Turning geospatial data into a usable marketplace journey
LandEye is an Earth observation marketplace where users discover satellite imagery, define an area of interest, inspect available products, and move toward ordering and payment. The central product surface is a map, but the engineering problem is not simply rendering tiles. Search criteria, spatial selection, imagery metadata, previews, availability, and commercial steps need to stay synchronized as the user changes context.
I joined as a senior frontend developer and focused on the interactive discovery and ordering experience. The product served an international audience and had to communicate data-heavy states without making the map feel like a specialist tool. My role included shaping reusable interface patterns, integrating remote data, and keeping multi-step workflows understandable across different screen sizes.
Building a map-led flow with clear state ownership
I built and refined workflows for imagery discovery, area selection, previews, ordering, and payment. A key part of the work was separating transient map interaction from committed search state. Users need to move, zoom, or adjust geometry without every gesture creating a confusing request or erasing the context they already established.
I represented remote operations with explicit loading, empty, error, and ready states. Imagery results combine visual footprints with metadata, so selection had to remain consistent across the map and catalogue views. Reusable components handled filters, result summaries, product information, and action states while keeping the domain-specific rules close to the workflow that used them.
The frontend also had to support a journey that moves from exploration into a transactional process. I kept ordering steps explicit and preserved the selected imagery context as users progressed, reducing the chance that a technically valid action felt disconnected from what they had chosen on the map.
Area-of-interest tools
Imagery discovery
Preview and metadata
Ordering workflow
Make complex state visible, not clever
Separate viewport from query
Share one selected result
Preserve context across steps
A clearer interface for data-intensive decisions
The delivered experience supports imagery search, area selection, product inspection, ordering, and payment as a connected product flow. I avoid assigning unverified conversion or performance numbers to that work. The concrete result is a production interface that makes specialized geospatial data usable through familiar interaction patterns and explicit system feedback.
This project strengthened my approach to map-heavy products: the map should not become the state manager for the entire application. Spatial interaction, remote query state, selected business entities, and transactional progress have different lifecycles. Keeping those lifecycles visible in the architecture makes the interface easier to reason about, test, and extend.