About Me

Hi, I’m Ramesh

Welcome to my corner of the internet. I am a curious person and a machine learning engineer. I like to build things. I want to solve real problems with computer vision and deep learning. Computer vision teaches computers to understand images. Deep learning is a kind of machine learning that uses large neural networks.

Right now, I lead machine learning work at Planet Labs. I build models that read satellite images. They track changes in the environment and map infrastructure around the world. I like to turn large, messy data into useful answers.

Before Planet, I worked at Siemens Corporate Research. I worked on visual anomaly detection and segmentation. Anomaly detection finds things that look unusual. Segmentation labels every pixel in an image. I have published at top conferences such as ICML, and I hold a patent in weakly supervised learning. That means training models with few or imperfect labels.

I earned my Master’s degree at Rochester Institute of Technology. Since then, I have worked where geospatial data, AI, and real-world impact meet. Geospatial data is data tied to places on Earth.

Outside of tech, I run, bike, and travel. I am curious about other cultures, and I like the stories I find in unfamiliar places.

This site shows my projects, research, and ideas. Explore it, and get in touch if something interests you. My links are in the footer.