Agronomist | Data Scientist | Founder
Bridging Biology and Software
Agronomist, Data Scientist, and Founder of Dakota AI. I build full-stack solutions that bridge the gap between biological data and actionable software.
About Me
My journey began in the fields as a Production Agronomist, managing variety trials and disease nurseries. I realized that the future of agriculture lies in data.
I am now pivoting to Data Science to build the tools that will drive the next generation of precision agriculture, utilizing my background in remote sensing, drone technology, and statistical modeling.
5+
Years Experience
Part 107
Remote Pilot
Education
MS in Data Science
North Dakota State University
Key Modules: Applied AI, Fundamentals of Data Engineering, Cloud Computing.
MS in Plant Science
North Dakota State University
Thesis: Importance of plant spacing heterogeneity in sunflower performance and phenotyping using remote sensing.
Crop and Weed Science, Agronomy
North Dakota State University
Summa Cum Laude | Minors in Botany and Biological Science.
Experience
Founder & Lead Data Scientist
Dakota AI LLC
Established a data consultancy to provide custom AI solutions for agricultural clients. Deployed commercial websites handling full lifecycle from DNS to frontend.
Production Agronomist
Southern Minnesota Beet Sugar Cooperative
Led sugar beet variety trials, coordinated disease nurseries, and managed a team to improve yield and disease resistance.
Graduate Research Assistant
USDA/NDSU
Researched sunflower plant spacing using UAS for canopy closure measurement. Presented findings at conferences.
Skills & Certifications
Data Science & Engineering
- Python (Pandas, NumPy, Scikit-Learn)
- Deep Learning (PyTorch, Keras)
- R (Statistical Computing)
- SQL & Cloud Computing
Geospatial & Remote Sensing
- ArcGIS Pro & QGIS
- Agisoft Metashape & Pix4D
- UAS Flight Operations
- FAA Part 107 Remote Pilot
Deployment & Web
- Full-Stack Development
- DNS & Hosting Management
- UI/UX Design
- Business & Client Relations
Agronomy & Research
- Experimental Design (RCBD)
- High-Throughput Phenotyping
- Official Variety Trials (OVT)
- SAS (GLM/MIXED Procedures)
Certifications
FAA Part 107 Remote Pilot
Case Studies
High-Throughput Phenotyping & Yield Prediction
The Problem: Determining if non-uniform plant spacing (skips/doubles) negatively affects yield and how to measure it efficiently.
The Tech Stack: UAV (MicaSense RedEdge MX), ArcGIS Pro, ImageJ, SAS.
The Work: Processed drone imagery to calculate canopy coverage using EVCI (Enhanced Vegetation Coloration Index) and stitched images to remove background noise for yield analysis.
View PublicationAutomated Sunflower Bloom Stage Detection
The Problem: Manual bloom staging is labor-intensive and prone to error.
The Tech Stack: Python, PyTorch/Keras, OpenCV, Computer Vision.
The Work: Developed a computer vision model to recognize different growth stages in a nursery setting, increasing data accuracy for breeding programs.
Predictive Disease Modeling for Root Rot
The Tech Stack: Statistical Modeling, Machine Learning (Scikit-Learn).
The Work: Analyzed 20 years of historical agricultural data to extract insights, demonstrating the ability to handle longitudinal datasets and perform complex data cleaning.
Ventures & Development
Dakota AI
My Consulting Firm. A specialized consultancy bringing artificial intelligence and data solutions to the agricultural sector in the Northern Plains.
Proves business viability, client acquisition, and full-lifecycle project delivery.
Visit DakotaAI.usFonicTonic
Full-Stack Playground. A creative web application demonstrating front-end design, audio interactivity, and direct-to-consumer productization.
Visit FonicTonic.comPublications
- Olson, N. A., Trostle, C., Meyer, R., & Hulke, B. S. (2024). Canopy closure, yield, and quality under heterogeneous plant spacing in sunflower. Agronomy Journal. [DOI Link]