Embedded Perception Engineer

About the role

As an Embedded Perception Engineer at Havoc, you will own the perception stack that enables autonomous surface vessels to sense and understand complex maritime environments in real time. You will design, deploy, and optimize vision and sensor‑fusion pipelines that run reliably on edge hardware under demanding conditions.

Key responsibilities

• Develop and maintain end‑to‑end perception pipelines deployed on operational systems.
• Design, train, and optimize vision models (e.g., ViT, CLIP, SAM) for maritime object detection, classification, and tracking.
• Fuse multi‑modal sensor data (camera, radar, lidar) and create low‑level sensor drivers.
• Deploy models with NVIDIA Triton Inference Server and optimize for latency, throughput, quantization, and pruning on embedded platforms.
• Integrate perception outputs into the broader autonomy stack and support field troubleshooting.
• Define metrics, monitor system health, and investigate perception failures.
• Collaborate with operators and customers to translate field challenges into engineering solutions.

Required profile

• Bachelor’s degree or higher in Computer Science, Electrical Engineering, Robotics, Machine Learning or a related field.
• 3+ years of experience developing and deploying computer‑vision or perception systems on Linux using C++ and/or Python.
• Hands‑on experience with modern deep‑learning frameworks and large vision models.

Required skills

• C++
• Python
• Linux development
• NVIDIA Triton Inference Server
• Vision models such as ViT, CLIP, SAM
• Sensor fusion (camera, radar, lidar)
• Model optimization techniques: quantization, pruning, latency/throughput tuning