Machine Learning Engineer III – Computer Vision & Data Systems

We are seeking a Machine Learning Engineer to build production-grade computer vision pipelines and manage end-to-end ML data infrastructure for autonomous aerial systems. You will own the full lifecycle from data curation through model deployment on edge hardware. 

Responsibilities: 

ML Model Development & Deployment 

• Design, train, and optimize object detection and tracking models (YOLO, RT-DETR, ViT) for aerial platforms 

• Deploy models on NVIDIA Jetson hardware using TensorRT, achieving real-time inference 

• Integrate vision models into ROS2 autonomy stacks with efficient message passing and synchronized sensor data 

• Profile and optimize model performance to meet compute, power, and latency constraints on embedded systems 

Data Infrastructure & MLOps 

• Build and maintain data pipelines for ingestion, labeling, versioning, and quality control of large-scale aerial imagery datasets 

• Design active learning loops and data selection strategies to maximize model performance with minimal labeling effort 

• Implement automated model evaluation frameworks and performance monitoring for deployed systems 

• Create synthetic data generation pipelines and data augmentation strategies to improve model robustness 

System Integration 

• Develop ROS2 nodes for real-time sensor processing, model inference, and downstream autonomy integration 

• Work with GStreamer and Jetson multimedia APIs for efficient camera pipeline management 

• Collaborate with perception and autonomy teams to ensure vision outputs meet system requirements 

Qualifications: 

• Bachelor's in Computer Science, Electrical/Computer Engineering, Robotics, or related field (Master's preferred) – equivalent industry experience also welcome 

• 4-6+ years of experience in machine learning with focus on computer vision applications 

• Strong proficiency in PyTorch and modern deep learning frameworks 

• Hands-on experience deploying models on NVIDIA Jetson platforms (Orin, Xavier, or similar) 

• Solid understanding of model optimization techniques: quantization, pruning, TensorRT, ONNX 

• Experience with ROS2 for robotics applications, including sensor integration and message handling 

• Proven track record building and managing ML data pipelines and infrastructure 

• Strong Python skills with experience in data processing libraries (NumPy, Pandas, OpenCV) 

• Familiarity with MLOps tools and practices (versioning, experiment tracking, CI/CD for models) 

Preferred: 

• Experience with C++ for performance-critical vision processing 

• Knowledge of CUDA programming or GPU optimization 

• Familiarity with data labeling tools and annotation management systems 

• Experience with synthetic data generation or domain adaptation techniques 

• Understanding of aerial imagery characteristics and challenges 

• Background in object tracking algorithms (Kalman filters, Hungarian assignment, multi-object tracking) 

• Experience with GStreamer, Jetson BSP, or embedded Linux development 

• Knowledge of model monitoring and performance debugging in production systems 

• Familiarity with DVC, MLflow, Weights & Biases, or similar MLOps platforms 

#LI-BC1