Projects
Engineering-focused work in geospatial systems, applied AI workflows and production-ready web platforms. Each project is presented as a system: architecture, constraints, deployment and usage.
← All projects
MSc dissertation modeling forest canopy height from PlanetScope high-resolution imagery using a stacking ensemble trained against GEDI spaceborne LiDAR.
Developed at NOVA University Lisbon (Electrical and Computer Engineering), the model combines 9 base regressors (Random Forest, XGBoost, LightGBM, Extra Trees, SVR, Ridge, Lasso, ElasticNet and KNN) through a Ridge meta-learner, trained on 99 spectral, texture and terrain features. Validated over two Portuguese forest areas (Monsanto and Serra da Lousã) against GEDI rh98 reference heights.
Tech Highlights
- Stacking ensemble of 9 regressors combined via a Ridge meta-learner, with a per-pixel uncertainty estimate.
- 99 input features spanning PlanetScope 8-band reflectance, GLCM texture and terrain derivatives.
- Trained and validated on GEDI spaceborne LiDAR (rh98) across two Portuguese forest study areas.
Model Performance
Architecture
- PlanetScope 8-band Surface Reflectance composites paired with GLCM texture and DGT LiDAR-derived terrain features.
- GEDI rh98 relative-height footprints (optimized) used as ground-truth training reference.
- 9 base regressors (RF, XGBoost, LightGBM, Extra Trees, SVR, Ridge, Lasso, ElasticNet, KNN) trained independently on 99 features.
- Ridge meta-learner stacks the 9 base predictions into a single final height estimate.
- Per-pixel uncertainty reported as the standard deviation across the 9 base predictions.
- Standalone inference script applies the trained ensemble to new co-registered raster stacks.
Python
scikit-learn
QGIS
Remote Sensing
GEDI LiDAR
PlanetScope
Ensemble Learning
Community-driven geospatial football mapping platform combining automated data scraping, interactive visualization and user-generated contributions.
Built with Python scraping pipelines, Zerozero.pt ingestion flows, Cloudflare D1 and modern web mapping technologies. Structured as a continuously updated product with contributor input and deployment-ready architecture.
Tech Highlights
- Automated ingestion pipeline for football clubs and location normalization.
- Data model designed for community submissions and moderation workflows.
- Geospatial rendering optimized for responsive and interactive usage.
Live Metrics
Architecture
- Python scraping pipeline orchestrates periodic source collection.
- Zerozero.pt ingestion stage validates and transforms raw entities.
- Cloudflare D1 persists club, location and contribution records.
- Map rendering stack serves responsive interactive geospatial views.
- Community contribution flow supports submission and review lifecycle.
- Deployment architecture targets global edge delivery via Cloudflare.
Python
JavaScript
Cloudflare D1
Geospatial
Web Scraping
ETL
Community Data
Personal portfolio platform featuring multilingual support, curl access, interactive 3D rendering and custom frontend tooling.
Custom frontend engineering with multilingual UX, real-time interactions and platform-oriented delivery patterns. Deployed globally using Cloudflare Pages.
Technical Highlights
- Multilingual content flow with dynamic language switching.
- Curl-compatible output and route behavior design.
- Interactive 3D rendering pipeline tuned for lightweight delivery.
JavaScript
HTML
CSS
Three.js
Cloudflare Pages
Multilingual
Curl UX
Open research archive documenting the life and sculptural works of itinerant Buddhist monk Mokujiki Shōnin (1718–1810), who carved over 1,000 devotional wooden sculptures across Japan.
Structured catalogue spanning 95 documented works across 63 sites and 17 Japanese prefectures, with 61 archived primary sources and scholarly papers. All materials released under CC0 for open academic and cultural use.
Highlights
- Catalogue of 95 works across 63 sites, indexed by prefecture and location.
- Primary source archive with 61 scholarly and historical documents.
- Community contribution model with open submission and CC0 licensing.
Portfolio Tracks
Live Projects
Production systems with active users, uptime constraints and deployment cadence.
Research & ML
Remote sensing and applied AI projects with large imagery previews, metrics panels and technical writeups.
Experimental
Feature experiments including comparison sliders, interactive maps and advanced UI prototypes.
Infrastructure
Cloud deployment, edge delivery, telemetry hooks and maintainable engineering workflows.