Computer Vision Engineer
JOB DESCRIPTION
1. JOB TITLE
Computer Vision Engineer 2D/3D Vision & LiDAR
2. JOB SUMMARY
We are looking for a Computer Vision Engineer with hands-on experience in building and deploying real-world
Computer Vision and LiDAR-based systems.
The candidate will work on projects involving cameras, LiDAR sensors, 2D/3D vision, object detection,
tracking, measurement, and sensor integration for industrial environments such as factories, warehouses,
logistics, and robotic systems.
The role involves working on the complete pipeline from sensor data acquisition and processing to
detection, measurement, calibration, 3D positioning, and real-time deployment on edge devices.
The candidate should be comfortable working with real-world and noisy sensor data and should have
experience working with multiple LiDAR sensors/technologies and their SDKs, drivers, and data formats.
3. KEY RESPONSIBILITIES
Develop Computer Vision solutions for image and video data, including object detection, classification,
tracking, segmentation, OCR, ANPR, and defect/anomaly detection.
Process and analyze LiDAR point clouds for filtering, noise removal, downsampling, ROI
extraction, ground/plane removal, clustering, object detection, and measurement.
Work with different LiDAR sensors, SDKs, drivers, and data formats.
Handle real-time sensor data streams, including data acquisition, timestamps, packet loss,
reconnection, and sensor health monitoring.
Perform camera calibration and LiDAR calibration, including intrinsic and extrinsic calibration.
Work with different coordinate systems such as camera, LiDAR, robot, machine, and world coordinates,
and perform transformations between them.
Develop Camera-LiDAR fusion solutions for localization, depth, and measurement.
Develop solutions for distance, depth, object position, dimensions, and spatial measurements using
cameras and LiDAR.
Train, evaluate, and improve Computer Vision deep learning models using appropriate metrics such
as precision, recall, mAP, localization error, and measurement accuracy.
Prepare, clean, annotate, and manage datasets for Computer Vision and LiDAR-based models.
Optimize Computer Vision and LiDAR pipelines for real-time performance on CPU/GPU and edge
devices.
Design reliable failure handling so that the system can identify poor data, occlusion, low
confidence, or sensor issues instead of producing unreliable results.
Integrate Computer Vision/LiDAR outputs with robotics, PLCs, and other software systems.
Support system testing, deployment, commissioning, troubleshooting, and performance optimization in real-
world environments.
Maintain clear technical documentation, test cases, and troubleshooting/commissioning documentation.
4. REQUIRED SKILLS & TECHNOLOGIES
Programming & Development
Strong Python programming skills.
Working knowledge of C++ is preferred, especially for performance-critical applications.
Good understanding of Git and Linux/Ubuntu.
Experience developing production-ready Computer Vision applications.
Computer Vision & Image Processing
Strong understanding of Computer Vision fundamentals.
Hands-on experience with OpenCV.
Image preprocessing, filtering, thresholding, edge detection, contours, morphology, feature matching, and
geometric operations.
Understanding of camera models, coordinate systems, transformations, and pose estimation.
Hands-on experience with PyTorch and/or TensorFlow.
Experience with object detection frameworks such as YOLO or similar.
Experience with: Object Detection, Image Classification, Segmentation, Object Tracking
Knowledge of tracking algorithms such as SORT, DeepSORT, ByteTrack, or similar.
Experience with model training, evaluation, optimization, and deployment.
Knowledge of ONNX and/or TensorRT is preferred.
OCR & Video Analytics
Experience with OCR technologies.
Experience processing real-time camera/video streams.
Understanding of video decoding, frame processing, event detection, and real-time video analytics.
2D/3D Computer Vision
Understanding of 2D-to-3D projection and 3D coordinate systems.
Knowledge of depth and distance estimation.
Experience with stereo cameras, depth cameras, or LiDAR for 3D perception.
Understanding of rotation, translation, transformation matrices, and coordinate-frame conversions.
LiDAR & Point Cloud Processing
Hands-on experience working with LiDAR sensors and point cloud data.
Experience with multiple LiDAR sensors/technologies and their SDKs or drivers.
Experience with: Point cloud filtering, Voxel/downsampling, Outlier/noise removal, ROI extraction,
Ground/plane segmentation, RANSAC, Clustering, 3D bounding boxes, Object measurement,
Point cloud registration/alignment.
Understanding of LiDAR characteristics such as range, intensity, field of view, noise, and scan
patterns.
Knowledge of point cloud registration techniques such as ICP and rigid transformation estimation.
Experience integrating cameras and LiDAR sensors.
Understanding of Camera-LiDAR calibration and sensor fusion.
Edge & Real-Time Deployment
Experience deploying Computer Vision solutions on NVIDIA Jetson or similar edge devices is preferred.
Basic understanding of CUDA and GPU-based inference.
5. REQUIRED EXPERIENCE
5 to 6 years of hands-on experience in Computer Vision, with at least 2 to 3 years working directly with
LiDAR.
Practical experience working with LiDAR and point cloud data.
Experience working with multiple LiDAR sensors/technologies and their SDKs, drivers, or data formats.
Experience developing and deploying at least one real-world Computer Vision, LiDAR, robotics, or
industrial vision system.
Experience with camera/LiDAR calibration and coordinate transformations.
Experience converting sensor measurements into real-world dimensions, positions, distances, or
coordinates.
Experience testing, debugging, and improving systems using real-world field data.
Bachelor's or Master's degree in Computer Science, Electronics, Robotics, Engineering, or a related
field, or equivalent practical experience.
6. GOOD TO HAVE SKILLS
Experience with ROS or ROS 2.
Experience with 3D deep learning and point-cloud models such as PointNet++, PointPillars, or similar.
Experience with SLAM, mapping, or localization.
Experience integrating Computer Vision/LiDAR with industrial robots such as FANUC, ABB..
Experience with multi-sensor time synchronization.
Experience with Docker and CI/CD.
Experience with industrial cameras and machine-vision systems.
Experience building simple monitoring dashboards or interfaces for live camera/LiDAR data.
Knowledge of industrial safety concepts.
Experience with logging, databases, and traceability systems.
7. IDEAL CANDIDATE PROFILE
The ideal candidate should be someone who can work across the complete Computer Vision + LiDAR
pipeline, from sensor data acquisition and preprocessing to AI inference, 3D perception, calibration, sensor
fusion, real-time optimization, and deployment.
The candidate should have a strong hands-on and problem-solving approach and be comfortable working
with real-world industrial conditions, noisy sensor data, hardware limitations, and field-deployment challenges.