# Perception Engineer

**Company:** [Origin](https://hotfix.jobs/companies/origin-tech)
**Location:** Bengaluru, India
**Role:** ML Engineering
**Skills:** Python, C++, Ros 2, PyTorch, TensorFlow, Pcl, Open3D, Lidar, Computer Vision, Point Clouds, TensorRT, Onnx, Ransac, Icp, Ceres
**Posted:** 2026-07-23

> Builds and deploys 3D perception systems for autonomous construction robots, combining cameras, LiDAR, deep-learning vision, and point-cloud geometry. The role requires Python and C++, ROS 2, computer vision, sensor fusion, and production optimization for edge hardware.

## Job Description

## Responsibilities
- Develop and deploy 3D perception components for geometric scene understanding using depth sensors, LiDAR, and RGB cameras.
- Build ROS 2 nodes that process and interpret point clouds, depth maps, and image streams for environment modeling and task planning.
- Train and integrate deep-learning models for 3D semantic understanding, including surface analysis and object segmentation.
- Design sensor-fusion strategies combining visual, inertial, and spatial data for scene reconstruction and robot localization.
- Benchmark and optimize perception models for edge-compute platforms such as NVIDIA Jetson using TensorRT or ONNX.
- Collect and curate real and synthetic datasets; automate training pipelines and experiment tracking.
- Collaborate with manipulation, navigation, and cloud robotics teams to deliver production-ready perception stacks for autonomous operation in dynamic construction environments.

## Requirements
- Strong understanding of linear algebra, probability, and geometry.
- Coursework or projects in computer vision or robotics perception.
- Proficiency in Python 3.x and C++17/20.
- Familiarity with Git and CI workflows.
- Experience with ROS 2, including `rclcpp`, `rclpy`, custom messages, and launch configurations.
- Familiarity with deep-learning vision using PyTorch or TensorFlow, including classification, detection, or segmentation.
- Hands-on experience with point-cloud processing using PCL or Open3D.
- Understanding of voxel grids, KD-trees, RANSAC, and ICP.

## Nice-to-Haves
- Camera–LiDAR calibration experience.
- Experience with real-time optimization libraries such as Ceres or GTSAM.

## Compensation and Benefits
- No compensation information provided.

## Similar jobs

- [AI Engineer - New Verticals](https://hotfix.jobs/jobs/4159c72e-536e-4211-969c-6bcb9ad805fd) - Protege - Remote
- [Agent Systems Engineer](https://hotfix.jobs/jobs/2b04a068-018b-4fe0-915f-4fdf80d4c582) - Adaption Labs - San Francisco, CA
- [Inference Performance Engineer](https://hotfix.jobs/jobs/7718c69a-6d49-445e-8c7c-840bd1b3f346) - Adaption Labs - San Francisco, CA
- [AI Transformation Engineer](https://hotfix.jobs/jobs/410f3e6a-4e24-4e94-aeb8-fb7b1dd2384f) - Dialpad - Bengaluru, India
- [Senior Software Engineer, NLP](https://hotfix.jobs/jobs/be3f0222-b08a-4620-9ddc-402209ea73f7) - GitLab - Remote

**Apply:** https://hotfix.jobs/jobs/427d13f8-7cdd-406a-a6b2-ce5059d745a9
**Canonical:** https://hotfix.jobs/jobs/427d13f8-7cdd-406a-a6b2-ce5059d745a9