BCNFS Talk: Scalable Geospatial Pipelines with GigaSpatial

Geospatial data is critical for global development, but practitioners constantly battle fragmented data, scaling issues, and complex integrations across vector and raster formats. Enter GigaSpatial, an open-source Python library built by UNICEF’s Giga initiative to help connect every school in the world to the internet.

In this talk, we will explore the architecture behind GigaSpatial and how it replaces the need to stitch together a dozen different tools to answer a single spatial question. We will dive into how the library orchestrates data collection, transformation, and standardisation pipelines at scale. By leveraging strict configuration validation, object-oriented data connectors, and seamless multi-store management, GigaSpatial provides a unified API to wrangle spatial data. Finally, we’ll look at real-world applications across 30 countries, from validating point-of-interest data against multiple global building footprints to running zonal aggregations using different grid systems.

Whether you are a data engineer, a GIS professional, or a Python developer interested in open-source architecture, you might find this talk an interesting look into how we build robust spatial pipelines; and we would love to explore ways you can contribute to this global mission.

Speaker: Utku Can Ozturk, Data Scientist and main developer of the Gigaspatial project

Date: 21 May
Start time: 19:00 h
Location: Akasha Hub - Carrer de la Verneda, 19, Nau 1 · Barcelona

Agenda:

• 19:00 h: Welcome

• 19:05 h: Talk

• 19:50 h: Networking