TixelJobs
J
Janicki Industriesvia muse

Robotics AI Engineer - Robot Learning

REMOTE$120K - $220K/yrPosted 5d ago
RoboticsMid LevelFull-time#remote

About the Role

**Help us teach robots to do things they've never done before.**


**Janicki Industries \| Sedro\-Woolley, Washington \| Full\-Time**


**Onsite, Hybrid and Remote opportunities available \| $120,000–$220,000 annually**



We're building robots that learn from people, practice in simulation, and develop the ability to perform useful work in the real world.



At Janicki Industries, we've spent more than 30 years solving difficult engineering problems and building advanced hardware for aerospace, defense, and space programs. Now we're bringing that experience to one of the most exciting challenges in engineering: creating intelligent robots that can adapt to new tasks and work in environments that were never designed for automation.



And we have some extraordinary opportunities to put that technology to work.


**A Different Kind of Factory**



Janicki has committed to building an $800 million advanced manufacturing facility in Montana to produce precision metallic components for the aerospace and defense industries. Ground has already been broken on the first building.



We're designing the facility around a high level of automation. The first phase is expected to include dozens of five\-axes milling machines, automated warehouses, robotic material loading, automated coordinate measuring machines, and automated metal finishing and coating systems. Massive robotic vehicles will move pallets weighing as much as 40,000 pounds throughout the factory.



But what makes this project particularly interesting isn't just the scale.



Most factory automation is designed to do the same thing, over and over, at high speed. Our challenges are different. We manufacture highly complex parts in relatively small quantities, often with significant differences from one job to the next.



This is high\-mix, low\-volume manufacturing, and it presents an entirely different set of robotics problems.



We need robots that can recognize unfamiliar parts, understand changing environments, adapt to different tasks, and work alongside sophisticated manufacturing equipment. Many of the capabilities we want simply don't exist as ready\-made solutions.



We see enormous potential in combining modern AI, computer vision, robot learning, and simulation to solve these problems. Our goal is to make this one of the most advanced and flexible manufacturing facilities in the world.


**You'll have an opportunity to help develop the technology that makes it possible.**


**Teaching Robots to Learn**



Our robotics team is exploring a new approach to automation: rather than programming every movement, we want to teach robots through human demonstrations, train neural networks to reproduce those behaviors, and use simulation to help them learn and improve.



We're interested in the rapidly developing world of vision\-language\-action models, imitation learning, reinforcement learning, and robotic foundation models such as NVIDIA Isaac GR00T.



We're also developing robots for applications outside traditional manufacturing. One of our first projects is an autonomous robot that cleans horse stalls using cameras, a robotic arm, and specialized tools. It's a deceptively difficult problem involving visual perception, object manipulation, and decision\-making in an unpredictable environment.



These projects share an important goal: creating robots that can perform useful physical work without having every action programmed in advance.



We've assembled a team of mechanical, electrical, and software engineers and are looking for an AI engineer who wants to help develop the intelligence behind these machines.


**What You'll Do**



You'll work on the technologies that allow robots to learn new skills and apply them to physical tasks.


* **Robot Learning:** Develop and train neural networks that translate visual information and human demonstrations into robotic actions. Work with imitation learning, reinforcement learning, and modern robot foundation models.
*
Share