D. Elliott

Atmospheric Data Science & Engineering

Dylan Elliott profile photo

Building atmospheric data and ML solutions for real-world systems

Hi! My name is Dylan.

I live at the intersection of atmospheric science, software development, and applied machine learning. My passions turn noisy remote-sensing and dynamical system data into robust predictions and production pipelines.

About Me

Atmospheric scientist working at SeekOps (Austin, TX), focused on developing drone-based CH4 and CO2 detection, quantification, and visualization methods using atmospheric transport physics, statistical modeling, and applied ML. Previously, was MS research graduate student in Atmospheric Science at Texas A&M (hybrid physics/ML weather prediction with data assimilation), also holds a BS in Earth and Planetary Science with Ocean concentration from UC Santa Cruz. Focused on bridging atmospheric physics with the engineering and high performance computing skills needed to make models and reliable data products.

Outside of work, I spend time surfing and building personal coding projects.

Core Skills

Domain Knowledge: numerical weather prediction, climate modeling, data assimilation, ensembles, atmospheric and oceanic transport modeling, signal processing, methane detection and quantification

Machine Learning + Data Wrangling: hybrid physics-ML methods via reservoir computing (echo state neural networks), computer vision modeling, training and evaluation, feature engineering, predictive modeling

Languages: Bash/Zsh, Python, Fortran 90, SQL, JavaScript, Excel

Data Engineering: Gitflow, ETL workflows/batch processing, reproducible analytics pipelines, tkinter

Cloud & Systems: AWS EC2, Azure, Linux, cron automation, HPC environments

Experience & Education

Click any entry to expand details. ● job  |  education

Projects

Click a folder to expand project details and file breakdown.

Papers & Publications

Peer-reviewed and preprint work.

Contact Me

If you'd like to get in touch, please reach out:

Resume: Available by request via email.