Founding Perception Engineer

Job Description

Develop the core perception stack for autonomous heavy machinery, transforming raw sensor data into spatial awareness. This founding role involves architecting robust 3D perception pipelines, fusing data from LiDARs, cameras, and GNSS/IMUs to power safety, teleoperation, and autonomy assist features for real-world construction sites.

Why this role is remarkable

• Fix humanity's heartbeat: construction, by building robots that re-wire the physical world and turn 50,000-pound machines into autonomous robots.
• Backed by Initialized Capital, Nebular VC, Ford Street Ventures, and 20+ founder-angels; achieved $220K ARR and a $2.5M pilot pipeline in under 3 months.
• Work directly in the field on live construction sites, deploying and iterating on cutting-edge autonomy systems, seeing your code move 20 tons of steel.

What you will do

• Architect and build robust 3D perception pipelines for teleoperated and semi-autonomous heavy machinery, fusing multi-modal sensor data.
• Develop real-time object detection, classification, and tracking systems to enhance operator situational awareness and enable autonomy.
• Optimize algorithms for real-time execution on embedded compute platforms like NVIDIA Jetson Orin, ensuring low latency and stability in challenging environments.

The ideal candidate

• A hacker obsessed with rapid iteration, prioritizing pragmatic, reliable algorithms before scaling to advanced learning-based methods.
• Former entrepreneur or someone with extreme agency, demonstrated by building projects from scratch, successful or failed.
• Experience developing and evaluating robust 3D perception systems, ideally for real-world robotic applications.