Applied ML & Data Science Research
We build ML systems grounded in data and validated by research. From satellite imagery to archival documents, we work where the data is messy and the impact is tangible.
Deep learning for environmental prediction, terrain analysis, and document understanding. From data collection to deployed inference.
Multi-source pipelines processing satellite imagery, LiDAR scans, climate reanalysis, and archival records into structured features.
GPU clusters, Kubernetes orchestration, and MLOps pipelines. We build and operate the infrastructure our models need.
Wildfire Prediction System
Domain: Environmental ML / Climate Science
Multi-source wildfire prediction for continental Portugal. Ensemble deep learning combining satellite imagery, weather reanalysis, terrain features, and 25 years of historical fire data to forecast risk across a 1km grid.
LiDAR Terrain Analysis Platform
Domain: Geospatial ML / Remote Sensing
Draw a polygon on the map, get 26 terrain analysis products back. LiDAR-powered pipeline processing Portuguese DGT data at 50cm and 2m resolution into elevation models, hydrology maps, landform classifications, and suitability analyses.
Document Intelligence Pipeline
Domain: Document AI / NLP / Computer Vision
AI-powered curation of archival records. Five-stage Airflow pipeline: OCR, layout analysis, named entity recognition, LLM-driven metadata enrichment, and multi-vector indexing. 11 distinct ML models orchestrated across GPU VRAM profiles.
Public Sector AI Services Cluster
Domain: MLOps / Infrastructure
End-to-end AI infrastructure for a Portuguese municipality. Kubernetes cluster with GitOps automation, conversational AI avatar for public service, and speech processing pipeline. Production-grade MLOps for local government.
We're always interested in problems where data meets real-world impact.
hello@neuralforge.pt