Senior Edge AI Perception Engineer

AION Robotics builds best in class autonomous vehicles designed for critical infrastructure monitoring, construction, mining and energy. Our vehicles operate in the toughest off-highway environments, where safety, accuracy, and efficiency are paramount. We are seeking a highly experienced GNC Engineer to design and optimize next-generation control architectures for our real-time edge autonomous systems.

About AION ROBOTICS CORPORATION

AION Robotics is a rapidly growing startup manufacturing advanced, rugged autonomous ground vehicles. We are bringing Industry 4.0 to outdoor commercial jobsites through the scaled deployment of turnkey autonomous products fully automating infrastructure monitoring and inspection tasks.

If our current open roles don't align with your background, we still want to connect with you. Join our Talent Network!

Description

Senior Edge AI Perception Engineer

Location: Arvada, Colorado

About AION Robotics

AION Robotics builds best in class autonomous vehicles designed for critical infrastructure monitoring, construction, mining and energy. Our vehicles operate in the toughest off-highway environments, where safety, accuracy, and efficiency are paramount.

Position Overview

We are seeking a highly skilled Edge AI Perception Engineer to design, optimize, and deploy cutting-edge deep learning models for real-time autonomous vehicle perception systems. This role involves building ultra-efficient pipelines for neural network inference on NVIDIA embedded platforms (Orin, Thor), integrating perception outputs with sensor fusion (LiDAR, camera), and pushing the limits of latency, throughput, and accuracy in real-world autonomous systems.

Key Responsibilities

• Neural Network Model Development & Optimization
• Build and manage optimized neural network pipelines tailored for edge deployment in autonomous vehicle systems.
• Implement, compress, and optimize models (pruning, quantization, scheduling) to run on GPU, DLA, and Tensor cores.
• Work with architectures including monocular depth models, YoloX, People Net, Res Net, and others.
• Pipeline & Resource Management
• Leverage NVIDIA Deep Stream, Tensor
  
  RT, CUDA, and TAO Toolkit to create high-performance perception pipelines.
• Manage model/hardware resource allocation across GPU/DLA for real-time scheduling and execution.
• Optimize pipelines "lens-to-detections" meeting ultra-low latency constraints on embedded devices such as NVIDIA Orin & Thor.
• Sensor Integration & Perception Enhancements
• Apply real-time geometric transforms to object detections and semantic segmentation results for geo-referencing and LiDAR point cloud filtering.
• CUDA accelerated 3D Terrain mapping
• Develop and maintain automated data collection pipelines, including dataset formatting, labeling workflows (CVAT), and fine-tuning for custom training.
• Camera Systems Engineering
• Implement real-time dewarped camera pipelines using GMSL drivers, VIC, and ARGUS APIs on embedded platforms.
• Collaborate with hardware engineers to achieve consistent calibration and synchronization across multiple sensors.
• Research & Innovation
• Stay up to date with bleeding-edge advancements in neural networks, edge AI optimization, and autonomous perception.
• Rapidly prototype and validate new models and methods for production deployment.

Qualifications

• Direct real-world experience in computer vision, deep learning, or edge AI systems.
• Strong proficiency with CUDA, Tensor
  
  RT, NVIDIA Deep Stream, TAO Toolkit, PyTorch/Tensor Flow.
• Demonstrated experience with model optimization techniques (e.g., pruning, quantization, distillation, scheduling).
• Hands-on experience deploying AI pipelines to embedded edge platforms (preferably NVIDIA Jetson Orin or Thor).
• Expertise in object detection, semantic segmentation, and depth estimation.
• Solid understanding of real-time embedded system constraints and low-latency optimization strategies.
• Strong programming skills in C++ and Python.

Preferred Experience

• Experience with low-level camera system integration, including:
• ISP tuning
• Intrinsic & extrinsic calibration algorithms
• Multi-camera synchronization and fusion
• Familiarity with sensor fusion pipelines combining camera, LiDAR, and IMU.
• Direct experience…