Senior Machine Learning Engineer

About the job

Senior Machine Learning Engineer – Robotics & Computer Vision

**** An innovative robotics company is seeking a Senior Machine Learning Engineer to help transform the future of warehouse automation through advanced AI and computer vision. This full-time opportunity focuses on developing and deploying next-Generation perception systems that enable robots to understand, interpret, and interact with complex physical environments. You'll work with cutting-edge technologies including Python, C++, PyTorch, YOLO, Mask R-CNN, Transformers, PointNet, TensorRT, ONNX Runtime, CUDA, and NVIDIA Jetson Orin to build high-performance 2D/3D vision models and deploy them directly onto robotic hardware.

Required Skills & Experience

• 5+ years of professional experience in Machine Learning and Computer Vision.
• Proven experience developing and deploying production ML solutions into physical-world applications such as robotics, autonomous vehicles, or IoT.
• Expert-level proficiency in Python and PyTorch.
• Working knowledge of C++ for model deployment, performance optimization, and system integration.
• Strong experience with 2D computer vision architectures, including YOLO, Mask R-CNN, and Vision Transformers.
• Experience with 3D computer vision, including PointNet, grasp generation, multi-view geometry, and camera calibration.
• Experience developing multimodal vision systems that combine 2D image inputs with 3D geometric information.
• Hands-on experience optimizing and deploying models using TensorRT, ONNX Runtime, CUDA, or similar technologies.
• Understanding of model quantization, graph optimization, and low-latency inference.
• Experience working with large-scale labeled datasets, including data curation, quality assurance, and identifying dataset bias.
• Experience developing and maintaining ML data pipelines and improving model performance through systematic evaluation.
• Ability to translate product requirements into engineering solutions and communicate technical trade-offs to non-technical stakeholders.

Desired Skills & Experience

• Experience deploying machine learning models on NVIDIA edge computing platforms, particularly NVIDIA Jetson Orin.
• Familiarity with Docker and containerized deployment environments.
• Experience with AWS or GCP, including services such as S3 and EC2.
• Familiarity with active learning strategies and automated dataset improvement workflows.
• Experience with data labeling platforms and annotation quality management.
• Familiarity with experiment tracking and ML lifecycle management tools.
• Experience integrating perception outputs with robotic control systems.
• Background in warehouse automation, robotic manipulation, industrial robotics, or autonomous systems.
• Experience with grasp planning, object pose estimation, and complex scene understanding.
• Experience mentoring engineers, conducting code reviews, and contributing to technical roadmaps.

What You Will Be Doing Tech Breakdown

• Programming: Python, C++
• Machine Learning: PyTorch
• 2D Computer Vision: YOLO, Mask R-CNN, Transformers
• 3D Computer Vision: PointNet, Multi-View Geometry, Grasp Generation
• Model Optimization: TensorRT, ONNX Runtime, CUDA
• Edge Computing: NVIDIA Jetson Orin
• Data Engineering: Dataset Curation, Active Learning, Data Quality Automation
• Cloud Infrastructure: AWS / GCP, S3, EC2
• Development & Deployment: Docker, ML Pipelines, Experiment Tracking
• Robotics: Perception Integration, Grasping, Scene Understanding

Daily Responsibilities

• Design, develop, and train advanced multimodal vision models that combine 2D image data with 3D geometry to solve complex robotic grasping and scene-understanding challenges.
• Architect and improve the company's perception stack to enable robots to operate reliably in dynamic and unpredictable warehouse environments.
• Own the end-to-end deployment of machine learning models, taking solutions from research and experimentation into production robotic hardware.
• Optimize model performance through graph optimization, quantization, and inference acceleration using TensorRT, ONNX Runtime, and CUDA.
• Deploy and integrate models onto NVIDIA Orin edge computing platforms, ensuring low-latency inference and efficient hardware utilization.
• Develop high-performance Python and C++ code to integrate perception outputs into the broader robotic control system.
• Own and improve existing labeled datasets and data pipelines, identifying bottlenecks and opportunities to improve data quality.
• Implement active learning strategies to identify challenging edge cases and systematically improve model robustness.
• Monitor dataset quality, detect statistical bias, and automate quality assurance throughout the ML pipeline.
• Conduct rigorous code reviews and establish best practices for production-grade machine learning software.
• Mentor junior engineers and contribute to the team's technical direction and long-term perception roadmap.
• Collaborate with robotics, software, and product teams to translate business requirements into scalable engineering solutions.
• Communicate technical decisions, model performance, and engineering trade-offs to cross-functional stakeholders.

****Applicants must be currently authorized to work in the US on a full-time basis now and in the future.