Podcast: Computer Vision for Business


Iu Ayala & Heather Couture on AI for Medical Imaging & Climate Tech


I recently had the pleasure of chatting with Iu Ayala on his Computer Vision for Business podcast.

We talked about:
✅ My journey from research to AI consulting
✅ Real-world applications of computer vision in pathology and environmental monitoring
✅ Key challenges and best practices for AI integration
✅ What makes a successful computer vision project

Be sure to check it out.

Research: Foundation Model for EO


Galileo: Learning Global and Local Features in Pretrained Remote Sensing Models


Imagine an AI that can read the Earth's story from space – pixel by pixel, month by month.

Remote sensing AI models vary widely in their capabilities. Gabriel Tseng et al. developed Galileo, a multimodal model designed to more comprehensively analyze Earth observation data.

The model integrates multiple data sources:
- Multispectral imagery from Sentinel-2
- Synthetic aperture radar (SAR) from Sentinel-1
- Elevation and land cover maps
- Time-varying weather data
- Static geospatial coordinates

Key technical features:
- Adapted Vision Transformer (ViT) architecture
- Processes 24 monthly time steps
- Analyzes 96 × 96 pixel images at 10m resolution
- Uses self-supervised learning to capture global and local features

In validation across multiple datasets, Galileo's three model variants performed consistently well. Ablation studies provided insights into the model's most critical characteristics.

The approach offers a more comprehensive method of analyzing satellite and geospatial data.

Blog: Foundation Models for Land and Carbon


What Are AI Foundation Models and How Are They Innovating Land and Carbon Monitoring?


Foundation models are now transforming many different imaging domains, and geospatial data is no exception.

This article by Zac Ogden and Craig Mills from Land and Carbon Lab outlines how foundation models are advancing land and carbon monitoring applications.

They can be fine-tuned for specific monitoring tasks and enable models to be trained with significantly less labeled data.

From mapping tree canopy heights to monitoring biodiversity to tracking restoration progress, foundation models are the building block enabling us to understand our planet in greater depth.

"But the true potential of foundation models lies in what comes next. AI is no longer just a tool for processing data, it’s a force for discovery. The biggest breakthroughs in Earth monitoring won’t come from what we already know, but from the new questions AI will help us ask — and the solutions it will help us find."