Note: The job is a remote job and is open to candidates in USA. Primis is a production-first leader in Embodied Intelligence, focusing on scaling physical AI in manufacturing. They are seeking a Senior ML Engineer specialized in Reinforcement Learning to architect a proprietary neural world model and develop policies for robots to learn and optimize manufacturing processes.
Responsibilities
- Define state, observation, action, and reward representations based on measurable manufacturing objectives; develop offline, model-based, or constrained RL methods
- Train and evaluate policies using learned world models, traditional simulation, and offline datasets to optimize across competing objectives like weld quality and cycle time
- Diagnose reward exploitation, unsafe behavior, and policy instability, translating research prototypes into dependable deployment systems
- Design methods that account for the messiness of the real world—specifically handling uncertainty, distribution shift, delayed outcomes, and sparse reward signals
Skills
- Master's or PhD in Computer Science, Robotics, Machine Learning, or related field, or equivalent practical experience
- Proven track record of developing and deploying reinforcement learning algorithms on real-world physical systems
- Proficiency in Python and deep learning frameworks (PyTorch or TensorFlow)
- Solid understanding of probability, statistics, and optimization
- Ability to navigate the 'bias toward shipping' from research to production
- Understanding of the current shift toward learning-based robotics
- Experience (or a strong interest) in training policies that handle complex, poorly modeled physics
- Proven ability to partner across functional teams to establish reliable evaluation methods
- Experience with simulation environments like MuJoCo or Isaac Gym
Benefits
- Fully remote (or onsite in Columbus, OH, if preferred).
- Full support for H1B transfers.
- Green Card sponsorship.
Company Overview