[Remote] Autonomy AI/MI Engineer
Note: The job is a remote job and is open to candidates in USA. Textron Specialized Vehicles Inc. is a global manufacturer of transportation and specialized equipment. The Autonomy AI/MI Engineer will develop and maintain intelligence, perception, and learning capabilities for autonomous systems, including data workflows, model training, evaluation, monitoring, and deployment. The role also supports continuous improvement through real-world and synthetic data, imitation learning, and reinforcement learning.
Responsibilities
- Own the intelligence of the autonomous system, ensuring that robust perception models enable reliable detection and understanding of the environment
- Develop automated data curation, labeling, and data quality workflows
- Evaluate and develop vision-language and foundation-model-based autonomy capabilities
- Develop model monitoring and drift detection approaches for deployed autonomy systems
- Evaluate AI model performance for robustness, explain ability, uncertainty estimation, and safety-critical deployment
- Maintain training pipelines effectively leveraging both synthetic and real-world data
- Establish a continuous improvement loop (data -> train -> deploy)
- Implement imitation learning (baseline driving) and RL-based policy optimization
Skills
- * Education: Bachelor's degree in Software or Computer Engineer, Information Technology, or related field required
- * Years of Experience: 2+ years of relevant experience required
- * Software Knowledge: Python, PyTorch/TensorFlow, Linux development environments, version control systems (Git) required
- * Required Skills: Strong experience in Python and ML frameworks (PyTorch / TensorFlow)
- Background in computer vision (object detection, segmentation)
- Experience working with real-world datasets and model training pipelines
- Understanding of model evaluation metrics and performance analysis
- Ability to iterate quickly on models using data-driven insights
- * 10-20% travel required to collect field data, evaluate perception performance, support data capture campaigns, and validate model performance in operational environments
- * Valid driver's license required
- * Preferred Skills: Reinforcement learning (PPO, DQN) and imitation learning
- Synthetic data and simulation-based training
- Sensor fusion (vision + LiDAR/radar)
- Deployment to edge platforms (TensorRT, NVIDIA stack)
- MLOps workflows and model lifecycle management
Benefits
- Will consider remote work for this opportunity. Employees who are working in remote roles will work primarily offsite (from home).
Company Overview
Company H1B Sponsorship