Physical AI Engineer

Title: Physical AI Engineer: Robotics and Simulation

Location: Waterloo, ON (onsite, occasional hybrid days)

Type: Full-time, 1-year contract · Start: early November

About Neatco

Neatco Engineering builds AI vision and robotics systems for manufacturing and recycling. Our IntelSort platform combines AI perception and robotic control to automate tasks that fixed-path automation can't handle.

The role

We're building systems where cameras and AI models tell robot arms what to do, first in simulation and then on real industrial hardware. You'll join a small team and take a working prototype further.

We care less about whether you've used our exact stack and more about how fast you learn. We expect you to use AI coding tools such as Claude, Copilot or Cursor to get productive in an unfamiliar codebase within days, and to check that what they produce actually works.

What you'll do

• Get our camera-to-robot prototype running, understand it, and extend it.
• Connect perception (object detection and depth) to robot motion in simulation.
• Bring realistic physics simulation and calibration into the loop.
• Move fast with AI tools, and use tests and checks to keep the system reliable.
• Help carry the work from simulation toward real robots.

What you bring

• A robotics foundation: coordinate frames, kinematics, and motion planning or control, from coursework, research, competitions or work.
• Solid Python, and comfort reading C++.
• Hands-on time with at least one robotics framework or simulator, such as ROS 2, Isaac Sim, Isaac Lab or Gazebo.
• Experience working with industrial robots such as FANUC, KUKA, Universal Robots (UR), Yaskawa or others.
• Comfortable working on Linux.
• Daily use of AI coding assistants, and the judgment to verify their output.
• Proof that you pick things up fast: a project, repo, video, hackathon or robotics competition.

Nice to have

• MoveIt or ros2_control.
• PLCs such as Allen-Bradley, Omron, CODESYS, Siemens or others.
• Depth cameras and camera calibration.
• Computer vision models such as segmentation or detection.
• Reinforcement or imitation learning.
• Docker.

Details

• Full-time, 1-year contract, onsite in Waterloo, with potential for an extension.
• Compensation is discussed at the first interview.

How to apply

Apply here: https://tally.so/r/8No0Zr. It takes about 10 minutes and asks for your resume as a PDF and a link to something you built.

We move quickly: a screening call, then an onsite technical assessment, then a final conversation with our CEO and CTO, usually within two weeks.

Neatco uses AI to help screen and assess applications. Every hiring decision is made by people.

Accommodations are available on request at any stage of the process.