Experience
Where I’ve worked and what I did there. For project details, methods, and outputs, see Research.
Projects Data Coordinator
National Center for Ecological Analysis & Synthesis, UC Santa Barbara
Curate and publish Arctic research data for the NSF Arctic Data Center and process permafrost datasets for the Permafrost Discovery Gateway.
- Direct the archival, curation, and publishing of 340+ TB of research data, from geospatial rasters & vectors to climate model output and survey data
- Build open-source R & Python tools and run parallelized workflows on UCSB’s HPC clusters
- Mentor a team of 5+ interns and teach 3 annual courses on reproducible & scalable research (Teaching)
Research Data Scientist
NASA Ames Research Center (via KBR Inc.)
Built semantic search and question answering tools for mission documents in the Intelligent Systems Division, with Nija Shi and Guillaume Brat.
- Search system used on 3 missions, covering 2,000+ project documents
- Improved correct retrieval by 30+% by fine-tuning BERT models on NASA phrases & acronyms
- Sped up document auditing by 3+ months
Related project: Semantic Search for NASA Mission Documents →
Research Assistant
Environmental Systems Dynamics Lab, UC Berkeley (advised by Laurel Larsen)
Developed physics-informed LSTM neural networks in PyTorch to forecast watershed discharge.
- Improved model accuracy by 5% by adding physics-based constraints
- Tested robustness across 30 watershed datasets in varying climates
- Co-authored posters at the AGU Fall Meeting (2021, 2022)
Data Science Intern
ClimateAi
Built a multi-model ensemble for short-term temperature and precipitation forecasts.
- Improved forecasts by 20% over the best vendor’s model
- Combined 6 data sources across 30 customer locations & 4 lead times
- Automated forecast pipelines in Google Cloud with Xarray & pandas
Education
B.A. Data Science, emphasis in Earth Science
University of California, Berkeley
Study abroad in Statistics & Mathematics at Lund University, Sweden.