Job Description:
We are seeking an AI Infrastructure / MLOps Engineer to design and maintain the data, training, and production pipelines that support machine learning and autonomy models across ROZOR robots.
You will develop systems for dataset generation, annotation workflows, simulation training, experiment tracking, and automated model deployment.
Your work ensures that AI models powering ROZOR robots can be trained quickly, validated reliably, and deployed safely at scale
Responsibilities:
- Build and maintain data pipelines for ingesting, labeling, storing, and processing robot datasets
- Develop annotation tools and workflows for perception and autonomy tasks
- Manage ML training pipelines, experiment tracking, and automated evaluation frameworks
- Implement MLOps practices for continuous training, testing, and deployment of ML models
- Integrate cloud infrastructure for scalable computation and distributed training
- Develop simulation pipelines for synthetic data generation and reinforcement learning
- Work with autonomy, perception, and embedded teams to define data needs and model deployment constraints
- Ensure reproducibility, versioning, and traceability of datasets and ML models
- Optimize inference packaging for deployment on edge compute (Jetson, ARM-based CPUs, GPUs)
- Monitor model performance and data drift across deployed ROZOR robots
Preferred Qualifications:
- Bachelor’s/Master’s degree in Computer Science, AI, Software Engineering, or related field
- 2+ years experience in ML infrastructure, MLOps, data engineering, or related domains
- Strong proficiency with Python and ML tooling (PyTorch/TensorFlow, ONNX, MLflow, etc.)
- Experience with cloud platforms (AWS, GCP, Azure) and distributed systems
- Familiarity with CI/CD pipelines, Docker, Kubernetes, or orchestration tools
- Experience designing data workflows and storage architectures
- Understanding of machine learning model training, validation, and deployment
- Strong debugging skills and attention to scalability and reliability