AI Embedded Engineer IV

At ASI
, we are revolutionizing industries with state-of-the-art autonomous robotics solutions. Within the fields of agriculture, construction, landscaping, and logistics, we deliver technologies that enhance safety, productivity, and efficiency. With our core values of Simplicity, Safety, Transparency, Humility, Attention to Detail, Autonomy and Growth guiding everything we do, we're shaping the future of automation in dynamic markets.

As an AI Embedded Engineer IV
, you own the layer where artificial intelligence (AI) meets the machine. Models that run fine on a workstation have to run on an NVIDIA Jetson bolted to a vehicle, reading Light Detection and Ranging (LiDAR) returns and depth camera frames, sharing compute and bandwidth with everything else on the platform, in dust and heat and vibration. Making that work is the job.

You take trained models, get them running in real time on vehicle compute, retune them when field performance drifts, and build the software that moves data between sensors, compute, and the robot itself.

The role is deliberately split between software and hardware. Most of your week is embedded software:
Robot Operating System (ROS) nodes on the vehicle, Python and C++ bridges between compute and robot control, runtime optimization, and debugging timing and synchronization problems across process and hardware boundaries. The rest is hands-on hardware. You will select and bring up sensors and compute, build the rigs that carry them, run the wiring and power, and fix electrical problems yourself rather than filing a ticket.

As a Level IV engineer within ASI's five-level engineering structure, you independently lead complex initiatives, influence the team's technical direction, and provide technical guidance to other engineers, with broader platform strategy remaining aligned with engineering leadership and Level V technical authorities.

• Deploy trained models to embedded Central Processing Units (CPUs) and Graphics Processing Units (GPUs) on vehicle compute, owning export, optimization, quantization, and latency budgets.
• Retrain and fine-tune existing models when field performance drifts, and validate the result on the vehicle rather than only on a benchmark.
• Build and maintain ROS nodes, topics, and interfaces running on the vehicle, and keep them stable under real workloads.
• Write the Python and C++ bridges that connect compute to robot control, sensor drivers, and the rest of the autonomy stack.
• Profile and tune runtime performance against real constraints, including compute headroom, memory, thermal limits, and power budgets.
• Bring up new sensors and compute hardware, including NVIDIA Jetson platforms, LiDAR units, and depth and vision systems, through provisioning, configuration, and calibration.
• Design and build sensor and data-collection rigs, covering sensor selection, mounting, wiring, power, networking, and onboard recording, then take them into the field.
• Debug hard cross-boundary problems spanning timing, synchronization, coordinate frames, machine motion, and compute limits.
• Fabricate and modify the mounts, brackets, and fixtures your hardware needs, and build and debug supporting wiring, harnesses, and small custom circuits.
• Characterize what you deliver honestly, including where it works, where it fails, and what conditions break it, so downstream teams know what they are inheriting.
• Document approaches, assumptions, results, and known limitations, and hand off work the production teams can build on with confidence.
• Provide technical guidance to other engineers on embedded deployment, sensor integration, and on-vehicle debugging.

• Bachelor's degree in Robotics, Computer Science, Computer Engineering, Electrical Engineering, Mechanical Engineering, or a related technical field.
• Substantial experience developing embedded, robotics, or autonomous system software. Graduate-level research experience in a relevant field counts toward this experience.
• Demonstrated experience independently taking complex embedded or AI integration work from concept to a working, evaluated system on real hardware.
• Advanced proficiency in C++ and Python.
• Experience deploying trained neural networks to embedded or production runtime environments, including model export and runtime optimization.
• Experience retraining or fine-tuning existing models and validating performance changes.
• Strong experience with Robot Operating System (ROS or ROS
  2) or comparable robotics middleware on real vehicles or robots.
• Hands-on experience bringing up embedded compute platforms such as NVIDIA Jetson, including provisioning, drivers, and configuration.
• Experience integrating LiDAR, depth cameras, or other vision systems, including calibration and data synchronization.
• Working understanding of coordinate systems, geometric transformations, camera models, and sensor timing.
• Experience with Linux, version control, automated testing, and containerized development.
• Genuine willingness to work hands-on with…