Edge CV & AI Intern — Field Hardware, Real Impact

ITERRA; San Juan (PR); On-site; USD 17,000 - 24,000; Full time

Benefits offered by this job

Real hardware from week one

Direct access to CTO

Field deployments to hardware

Portfolio of work to show

Path to full-time position

Job summary

ITERRA is building a passive sensing stack for low-altitude airspace with edge-native computer vision on NVIDIA Jetson. You will develop fast, small models for detection and tracking of airframes in visible and IR, and push them to the field on rugged hardware in Puerto Rico.

join a small, high-trust team; you will own end-to-end components from data collection to deployment, including dataset curation, model evaluation, and integration with RF sensors.

Qualifications

• Pursuing or holding a B.S. in Computer Science, Computer Engineering, or a related field. This is an early-career seat and we expect to teach you things.
• Hands-on work with thesis, senior design, student competition, personal project, or bench experience counts.
• You have trained a model and evaluated its failures with clear reasoning.

Responsibilities

• Develop edge computer vision capabilities for Group 1-3 UAS detection and tracking on NVIDIA Jetson.
• Quantize models and use TensorRT for real-time performance on the device.
• Build and manage dataset labeling, versioning, and training/validation splits.
• Collaborate with RF and software teams to fuse detections into a single track.
• Create scripts, dashboards, and tooling to repeat experiments and demonstrations.

Skills

Python

Linux

Computer Vision

PyTorch/TensorFlow

C++

ROS

Edge deployment

Education

B.S. in CS/CE

Tools

TensorRT

ONNX

Job description

ITERRA is building a passive sensing stack for low-altitude airspace with edge-native computer vision on NVIDIA Jetson. You will develop fast, small models for detection and tracking of airframes in visible and IR, and push them to the field on rugged hardware in Puerto Rico.

join a small, high-trust team; you will own end-to-end components from data collection to deployment, including dataset curation, model evaluation, and integration with RF sensors.