I’m a research engineer supporting innovative work on multimodal foundation models, multidimensional data pipelines, and deep learning for CV/NLP. Most of my work until now has been in the geospatial and climate domains, though I’m excited about the emergence of AI reasoning agents.

My areas of focus have included ML weather forecasting, global map data generation, digital twins, 3D photogrammetry, newspaper OCR, medical imagery analysis, geographic bias in vision models, multilingual AI, and crowdsourcing.

I’ve worked in academic research environments, R&D labs, startups at various stages including as the first hire, B corporations, and tech nonprofits. I have a Master’s in Data Science and a Bacherlor’s in Networked and Social Systems Engineering from the University of Pennsylvania.

Always open to a new challenge, please reach out!

Research

Global atmospheric data assimilation with multi-modal masked autoencoders | Preprint 2024

Design Considerations for High Impact, Automated Echocardiogram Analysis | ICML 2020 Machine Learning for Global Health Workshop

Learning Word Ratings for Empathy and Distress from Document-Level User Responses | International Conference on Language Resources and Evaluation (LREC) 2020

Mitigating Geographic Bias of Image Classifiers with Multilingual Image Data | Bloomberg 2019 Data for Good Exchange

Detecting Roads from Satellite Imagery in the Developing World | CVPR 2019 Computer Vision for Global Challenges Workshop

Generating a Training Dataset for Land Cover Classification to Advance Global Development | NeurIPS 2018 Workshop on Machine Learning for the Developing World

Decoded: Crowdsourcing Summaries of Internet Privacy Policies | HCOMP 2017 Poster

Blogs

Announcing Helios, a satellite based solar analysis and forecast | Zeus AI blog

EarthNet: A multi-modal foundation model for global data assimilation of Earth observations | Zeus AI blog

Understanding Overture’s Global Entity Reference System | Overture Maps Foundation blog

Using Computer Vision to Create A More Accurate Digital Archive | NYTimes R&D blog

3D Satellite Maps: Updating 135 million km² of high-resolution imagery + launching GL JS v2 | Mapbox blog

30+ meter imagery update: improved recency, color, and resolution for seamless 3D terrain | Mapbox blog

Updated Aerial Imagery in the US | Mapbox blog

Talks

Generating Elevation Data from Stereo Pairs of Satellite Imagery | GeoPython 2023

Depth Perceptions - Perspectives on 3D & Digital Twins | SatSummit 2022

Mapping Historical “Street View” Images of New York City: Visualizing Geotagged Archival Photos | FOSS4G 2022

Under-Mapped Spaces: New Methods and Tools for Critical Storytelling | Stanford 2022

Deep Transfer Learning for Land Cover Classification on Open Multispectral Satellite Imagery | FOSS4G 2019

Last Update: August 2025