Perception Engineer

About Amidon Heavy Industries

Amidon Heavy Industries is a venture-backed startup transforming subsea inspection and monitoring. We build autonomous surface vessels (USVs) that autonomously launch and recover remotely operated vehicles (ROVs) to perform high-frequency inspections of subsea pipelines, telecom cables, and offshore infrastructure. Our system operates at a fraction of the cost of traditional crewed vessels, unlocking scalable, real-time subsea awareness for the first time. You can read more on our website: https://amidonheavyindustries.com/

The Role

We are hiring a Perception Engineer to own the perception and mapping stack for our USV-ROV system. You’ll develop the pipelines that turn raw subsea sensor data into actionable insights, enabling digital twins, anomaly detection, and real-time awareness for our customers. This is a foundational role where your work will directly shape both the product and the company.

Responsibilities

• Develop sensor fusion pipelines for sonar, video, IMUs, and navigation data.

• Build real-time mapping and reconstruction systems (SLAM, structure-from-motion, multi-view geometry).

• Develop perception algorithms for cable and pipeline detection, tracking, and condition monitoring.

• Integrate perception with mission planning and operator dashboards (Helm platform).

• Run simulations and at-sea tests, analyzing field data to refine algorithms.

• Ensure systems perform in real-world ocean conditions (noise, poor visibility, dynamic currents).
  

Qualifications

• 3+ years of experience in robotics perception, computer vision, or autonomy.

• Strong skills in C++ and Python.

• Experience with SLAM, sensor fusion, and real-time perception pipelines.

• Hands-on with ROS/ROS2 and robotic software frameworks.

• Familiarity with sonar, video, and navigation sensors.
  

Bonus Points:

• Experience in marine robotics or underwater perception.

• Background in deep learning for perception (segmentation, anomaly detection).

• Field robotics experience, building systems that survive outside simulation.