I work on applied machine learning for scientific problems: models that fuse messy, multi-modal real-world data into predictions scientists can use. At the NMSU KDD Lab, my senior-thesis research uses multi-branch LSTMs to predict rangeland herbaceous biomass and how it changes, fusing satellite, precipitation, cattle-GPS, and land-cover signals. At NASA's Planetary Data System (Atmospheres Node), I develop ELSA, the web application planetary scientists use to generate PDS4 metadata for large datasets and prepare it for archiving.
Selected work
Research
Predicting Rangeland Herbaceous Biomass with Multi-Branch LSTMs
KDD Lab, New Mexico State University · Sep 2025 – Present
Read the write-up
Multi-branch LSTMs that fuse satellite biomass history, on-ranch rainfall, collar-derived grazing pressure, and static land cover to predict herbaceous biomass and its period-over-period change on two grazed pastures, raising biomass-change R² from 0.36 to 0.93.
- LSTM
- multi-modal fusion
- temporal encoding
- remote sensing
- time-series regression
Engineering
ELSA: Metadata Tooling for NASA's Planetary Data System
Research Assistant / Software Engineer · May 2024 – Present
Read the write-up
A web application that walks planetary scientists through generating PDS4 metadata for large scientific datasets and preparing it for submission to NASA's PDS Atmospheres Node, including a new service for planetary model output.
- Python 3.11
- Django 4.2
- MariaDB
- xarray / netCDF4
- Bootstrap 5
- Gemini API
Publications & Presentations
Neakrase, L. D. V.; Chanover, N.; Hartwick, V. L.; Huber, L.; Ajo-Montaño, S.; Dey, R.; Miller, N. The PDS Atmospheric Modeling Annex: Adapting to the Community Need for a Planetary Atmospheres Modeling Output Repository. AGU Fall Meeting 2026, San Francisco, CA, 2026.
poster · in-review
Neakrase, L. D. V.; Chanover, N.; Hartwick, V. L.; Huber, L.; Ajo-Montaño, S.; Dey, R.; Miller, N. The PDS Atmospheric Modeling Annex: A new repository for modeling output. 58th Annual Meeting of the AAS Division for Planetary Sciences (DPS 58), Spokane, WA, 2026.
poster · accepted
Projects
pantry.ai
1st Place, Desert Dev Lab HackathonBuilt at a hackathon, with Christopher Meraz and Antonio Luna · Apr 2026
Installable PWA that reads a pantry from a photograph, tracks what is about to expire, and plans meals around it. Team build; I owned meal planning and the shopping list engine.
- Role
- Meal planning and shopping list; co-built the kitchen wall UI
- Stack
- Next.js
- TypeScript
- SQLite
- Drizzle ORM
- Google Gemini API
- ZXing
Read the write-up
Course project, CSCI 4425, with Mario Saenz, Asim Upreti, and Demetrius Billey · Feb 2026 – May 2026
Five model families compared on 4-class brain tumor MRI, from hand-crafted features to a Vision Transformer. Checking the data turned up 7% of the test set duplicated from training, and a conditional GAN that had collapsed to four fixed images.
- Role
- The Vision Transformer, the evaluation harness, and the GAN augmentation
- Stack
- Python
- PyTorch
- ResNet-50
- ViT-B/16
- Scikit-learn
- Streamlit
Read the write-up
Team project, Introduction to Data Mining · Oct 2025 – Dec 2025
Five regressors compared on 232,000 US Airbnb listings, with decision-tree feature selection cutting 1,475 columns down to the ten that carry 83% of the signal.
- Role
- Feature selection, the KNN and random forest models, and the comparison
- Stack
- Python
- Pandas
- Scikit-learn
- Matplotlib
Read the write-up
Individual project across seven graded milestones, CS 370 · Sep 2024 – Dec 2024
Compiler for J, a small imperative language designed for the course, written in C with Flex and Bison. Parses to an AST and emits RV32 assembly with working stack frames, function calls, and recursion.
- Role
- Grammar, scanner, and code generator
- Stack
- C
- Flex
- Bison
- RISC-V Assembly
- RARS
Read the write-up
News
pantry.ai won 1st place of 27 teams at the statewide Desert Dev Lab Hackathon.
PressNMSU Newsroom · Las Cruces Bulletin · KTSM · AOL
hackNMSU, NMSU's first hackathon, brought together 55 participants across 18 teams.
PressNMSU Newsroom · Organ Mountain News