Open to new roles
02 · Agtech robotics · sensing + data quality
Robotics co-op · 2025
Making field sensing and visual-data workflows hold up outside ideal conditions.
Robotics co-op work connecting field hardware, ROS2/OpenCV sensing, GPS metadata, and representative data curation for precision agriculture.
Role
Robotics co-op · 2025
What I built
Integrated thermal/GPS field sensing in C++ and Python using ROS2, OpenCV, NVIDIA Jetson, and MicroROS.
Selected evidence
Thermal + GPS field sensing
The problem
The challenge was not a clean lab demo: it was translating farmer feedback into technical requirements, then making sensing and downstream computer-vision data useful under field conditions.
Technical ownership
- 01Integrated thermal/GPS field sensing in C++ and Python using ROS2, OpenCV, NVIDIA Jetson, and MicroROS.
- 02Redesigned a multi-camera fixture to add thermal imaging and translated farmer feedback into technical requirements.
- 03Used ROI filtering, detector embeddings, UMAP, HDBSCAN, and grouped splits to create representative annotation and evaluation batches from a 150,000+ image field dataset.
System architecture
Multi-camera + thermal fixture → Jetson/OpenCV + ROS2/MicroROS capture → GPS-linked image sets → data-curation workflow.
Evidence + outcome
Tools + systems
In the work
What the work actually looked like.
The hardware, measurements, and interfaces behind the short version.
