Arun Kothari

Responsibilities:

• Led a Unity AR/VR simulation for Navy MRO—secured $2M and reduced setup time by 40%.
• Architected Perforce workflows and a SQL asset system, cutting integration issues by 70%.
• Built point cloud editing + clash detection tools to accelerate inspection and maintenance iteration.
• Implemented ICP + RANSAC registration for precise multi-sensor alignment on large-scale ship datasets.
• Developing a neural network–driven UMV control system to replace heuristics and improve autonomy.

Responsibilities:

• Designed and trained a UNet segmentation model to track oxygen atom movement in CeO2 reactions—cut annotation effort by 60%.
• Processed high-resolution microscopy data using deep learning segmentation to improve atomic-scale feature accuracy.
• Built an interactive React + p5.js visualization platform to make nanoparticle research accessible to students.
• Delivered weekly technical briefings on model performance and experiments to accelerate research decisions.