01 / FOCUS

Four disciplines,
one trajectory.

XR, AI and data engineering, applied to one domain: space and geospatial data. Immersive interfaces to explore it, machine learning to understand it, solid pipelines to make it usable.

A · XR

Extended Reality

Immersive interfaces for real use, not trade-show demos: spatial data workspaces, training simulators, operator tooling. Real-time 3D, hand tracking, physics, and interaction design that turns complexity into clarity. Especially powerful when the subject is a planet.

Unity / UnrealWebXRReal-time 3D Hand trackingInteraction designSpatial data viz in VR
B · AI

AI & Machine Learning

From notebook to production without losing the plot: computer vision on satellite imagery, anomaly detection, predictive models. Honest about accuracy/latency trade-offs, the best model is the one that actually runs where it's needed.

Computer visionPyTorchEO imagery analysis Anomaly detectionMLOpsEdge deployment
C · GEO

Space & Geospatial Data

Satellite imagery and Earth observation: ingestion of raster and multispectral data, reprojection, mosaicking, cloud masking, the journey from raw scene to analysis-ready layer. Then GIS: spatial queries, change detection, territory monitoring, cartography that non-experts can read.

Sentinel / CopernicusGDAL / rasterioCOG / STAC PostGISQGISGeoPandasChange detection
D · DATA ENG

Data Engineering

The infrastructure underneath everything: ETL/ELT pipelines, orchestration, cloud storage, data modeling. With a bias for the parts that matter in production, schema discipline, testing, observability, failure handling, because pipelines should be quiet at 3 a.m. and trustworthy at 9 a.m.

PythonSQLAirflowETL / ELT Cloud (AWS/GCP)Data modelingCI/CD