Description

In this role, you will focus on developing advanced perception algorithms for autonomous robots, optimizing models for various data types, and collaborating with teams to enhance AI infrastructure and workflows.

Requirements

• •Degree in Computer Science, Robotics, or related technical field
• •2+ years of professional experience in computer vision, deep learning, or robot perception
• •Strong understanding of Vision Transformers (ViTs), Convolutional Neural Networks (CNNs), and DETR-based models for object detection
• •Hands-on experience with frameworks including PyTorch, OpenCV, MMDetection, and/or Detectron2
• •Applied experience in Optimization & Deployment with TensorRT, ONNX, Docker, and/or NVIDIA DeepStream
• •Direct experience with Data Annotation using CVAT, and/or Label Studio, and/or 3D Point Labeler
• •Practical experience in Versioning & MLOps using GitLab CI/CD, MLflow, n8n, AWS SageMaker, and/or Bedrock
• •Strong programming skills in Python, C++, Bash
• •Strong analytical, problem-solving, and growth mindset
• •Excellent collaboration between cross-functional AI and robotics teams
• •Passion for innovation, autonomy, and pushing the boundaries of perception systems
• •Ability to work independently and manage multiple priorities in a fast-paced environment
• •Fluent in English
• •Experience with LLMs, VLMs, VLAs and multi-modal embeddings
• •Familiarity with semantic scene graphs-based 3D Gaussian splatting
• •Exposure to cloud-based data services or AI workflow orchestration
• •Contributions to open-source projects in AI/robotics or publications in the field

Tasks

• •Develop and deploy vision-based perception algorithms for autonomous inspection robots
• •Design and implement supervised, self-supervised, and unsupervised learning methods for multi-modal data
• •Build and optimize models for object detection and tracking
• •Implement anomaly detection techniques
• •Conduct LiDAR-based semantic segmentation and 3D point cloud understanding
• •Optimize models for on-device inference through quantization, pruning, and distillation
• •Deploy models on edge hardware like NVIDIA Jetson Xavier/Orin/Thor
• •Integrate perception pipelines into robotic systems using ROS1/ROS2
• •Collaborate with software and robotics teams
• •Support MLOps workflows for reproducible training and evaluation of AI models
• •Maintain cloud-based AI pipelines on AWS for scalable training and inference
• •Contribute to research on multi-modal and 3D representations
• •Participate in the continuous improvement of AI infrastructure for mission planning and robot autonomy

Tools & Technologies

Benefits