01 — R&D Leadership
We design and build the data systems, processing pipelines, and analytical frameworks that turn raw observation data into usable science — removing the engineering bottleneck so researchers and operators can focus on what they do best.
The situation we walk into
Observation data is being collected — acoustic surveys, sensor streams, satellite feeds, vessel logs — but the pipeline between raw data and usable output doesn't exist. Scientists spend months on data engineering that isn't their craft. Operators can't act on data they can't read. Institutions sit on datasets that are technically inaccessible.
The technology is available. The data is there. What's missing is the structure that connects them.
What we deliver
In practice
Cloud-based acoustic data-processing platform for the European Pelagic Freezer-trawler Association (PFA). Built on Azure with modular pipelines for ingest, processing, and analysis — capable of reprocessing ~40 TB of echosounder data in hours. Removes the data-engineering bottleneck so researchers focus on science rather than infrastructure.
Multi-partner industry R&D project with WMR and IMR. Partners collect and analyse commercial echosounder data from the European pelagic fleet to improve understanding of fish stocks and deliver decision support to skippers, fleet managers, and fisheries analysts. Includes ML-based species classification, a Coexistence Index for bycatch prediction, and an open-source method that reprocesses a decade of acoustic data (~40 TB) in hours rather than months.
Won via competitive tender: an interactive R-Shiny and Power BI data portal for the Pacific Community's FAME division, giving Pacific island member countries direct access to fisheries catch data for management decisions. Delivered full technical documentation and extended into a second phase in 2025.