Autonomous Navigation And Perception Engineer

Optm Media Solutions; Pune District, Bengaluru; In loco; INR 1.200.000 - 1.800.000; Tempo pieno

Descrizione del lavoro

Optm Media Solutions is seeking an Autonomous Navigation and Perception Engineer in Pune to develop real-time perception systems with deep learning models, sensor fusion, and optimized deployment on edge platforms.

The role requires strong Python and C++ (C++17), MATLAB/Simulink, and ROS 2 experience, with hands-on work in LiDAR/radar data processing and multi-modal fusion for automated navigation.

Competenze

• 3 to 6 years of hands-on experience in Computer Vision, Deep Learning, or Perception Engineering for real-time systems.
• Strong proficiency in deep learning frameworks and deployment tools, including PyTorch, TensorRT, and ONNX.
• Practical experience with model optimization techniques such as INT8 quantization, pruning, layer fusion, and inference optimization.
• Strong understanding of image processing, point-cloud processing, object detection, and tracking algorithms.
• Hands-on experience with ROS 2, preferably Humble, and familiarity with LiDAR processing, radar-based object tracking, and multi-sensor fusion.
• Strong programming skills in Python and modern C++ (C++17 or later).
• Experience with MATLAB/Simulink for algorithm development, simulation, validation, and performance analysis.
• Understanding of real-time performance evaluation using mAP, IoU, FPS, and latency metrics.

Mansioni

• Design, develop, and deploy real-time computer vision and perception algorithms for object detection, classification, tracking, and scene understanding.
• Develop and optimize deep learning models using PyTorch, ONNX, and TensorRT for real-time inference.
• Implement model optimization techniques, including INT8 quantization, pruning, layer fusion, and inference acceleration.
• Process and analyze real-time image, LiDAR point-cloud, and radar data for object detection and tracking.
• Develop multi-modal sensor fusion algorithms integrating camera, LiDAR, and radar data.
• Evaluate model performance using metrics such as mAP, IoU, FPS, inference latency, and accuracy.
• Work with ROS 2 (preferably Humble) for perception pipeline development, integration, testing, and deployment.
• Use MATLAB/Simulink for algorithm development, simulation, modeling, validation, and testing of perception systems.
• Profile and debug perception pipelines to improve computational efficiency and real-time performance.
• Collaborate with cross-functional teams to integrate perception solutions into production systems.

Conoscenze

Computer Vision

Deep Learning

Perception Engineering

PyTorch

ONNX

TensorRT

ROS 2 (Humble)

Python

C++17

MATLAB/Simulink

Strumenti

CUDA

CVAT/SuperAnnotate

RViz/rosbag

Descrizione del lavoro

Job Description

Experience: 3 to 6 Years
Job Title: Autonomous Navigation And Perception Engineer
Employment Type: Full-Time

Job Summary

We are looking for a skilled Autonomous Navigation And Perception Engineer with hands‑on experience in developing, optimizing, and deploying deep learning models for real‑time perception systems. The ideal candidate should have strong expertise in computer vision, sensor fusion, LiDAR and radar‑based object tracking, and model deployment on optimized inference platforms. Experience with MATLAB/Simulink, ROS 2, and modern C++ is highly desirable.

Key Responsibilities

• Design, develop, and deploy real‑time computer vision and perception algorithms for object detection, classification, tracking, and scene understanding.
• Develop and optimize deep learning models using PyTorch, ONNX, and TensorRT for real‑time inference.
• Implement model optimization techniques, including INT8 quantization, pruning, layer fusion, and inference acceleration.
• Process and analyze real‑time image, LiDAR point‑cloud, and radar data for object detection and tracking.
• Develop multi‑modal sensor fusion algorithms integrating camera, LiDAR, and radar data.
• Evaluate model performance using metrics such as mAP, IoU, FPS, inference latency, and accuracy.
• Work with ROS 2 (preferably Humble) for perception pipeline development, integration, testing, and deployment.
• Use MATLAB/Simulink for algorithm development, simulation, modeling, validation, and testing of perception systems.
• Profile and debug perception pipelines to improve computational efficiency and real‑time performance.
• Collaborate with cross‑functional teams to integrate perception solutions into production systems.

Required Skills and Qualifications

• 3 to 6 years of hands‑on experience in Computer Vision, Deep Learning, or Perception Engineering for real‑time systems.
• Strong proficiency in deep learning frameworks and deployment tools, including PyTorch, TensorRT, and ONNX.
• Practical experience with model optimization techniques such as INT8 quantization, pruning, layer fusion, and inference optimization.
• Strong understanding of image processing, point‑cloud processing, object detection, and tracking algorithms.
• Hands‑on experience with ROS 2, preferably Humble, and familiarity with LiDAR processing, radar‑based object tracking, and multi‑sensor fusion.
• Strong programming skills in Python and modern C++ (C++17 or later).
• Experience with MATLAB/Simulink for algorithm development, simulation, validation, and performance analysis.
• Understanding of real‑time performance evaluation using mAP, IoU, FPS, and latency metrics.

Good‑to‑Have Skills

• Experience with CUDA programming and GPU acceleration.
• Familiarity with CVAT or SuperAnnotate for dataset annotation and management.
• Experience with RViz, rosbag, GStreamer, and performance profiling tools.
• Knowledge of automotive perception systems, ADAS, autonomous driving, or robotics.
• Familiarity with model deployment on embedded GPU platforms and edge computing devices.
• Experience with MATLAB toolboxes for computer vision, sensor fusion, or automated driving.