Perception Engineer

Origin; Bengaluru; On-site; INR 1,500,000 - 3,000,000; Full time

Job summary

Origin is seeking a Perception Engineer to enable autonomous drywall-finishing robots to see and understand job-sites. You will design perception pipelines using camera, LiDAR fusion and 3D geometry for robust site awareness.

You will train models, perform sensor calibration, and optimize perception stacks for edge devices, collaborating with manipulation, navigation and cloud teams on production-ready systems.

Qualifications

• Solid grasp of linear algebra, probability and geometry.
• Proficient in Python 3.x and C++17/20; comfortable with git and CI workflows.
• Experience with ROS 2 (rclcpp / rclpy) and custom message / launch setups.
• Familiarity with deep-learning vision (PyTorch or TensorFlow).
• Hands-on work with point-cloud processing; know when to apply voxel grids, KD-trees, RANSAC or ICP.
• Exposure to sensor calibration algorithms.

Responsibilities

• Develop and deploy 3D perception components for geometric scene understanding using depth sensors, LiDAR, and RGB cameras.
• Process and interpret spatial data (point clouds, depth maps, image streams) for environment modeling and task planning.
• Train and integrate deep-learning models for 3D semantic understanding and quality evaluation.
• Design robust sensor fusion strategies that combine visual, inertial, and spatial data for scene reconstruction and robot localization.
• Benchmark and optimize perception models for deployment on edge compute platforms (e.g., NVIDIA Jetson) using TensorRT or ONNX.
• Collect and curate high-quality datasets (real and synthetic); automate training pipelines and experiment tracking.
• Perform and validate sensor calibration (camera, LiDAR, IMU) and develop calibration quality checks.
• Collaborate across robotics teams to deliver production-ready perception stacks for autonomous operation in dynamic construction environments.

Skills

Linear algebra

Probability

Geometry

Python

C++

ROS 2

Deep learning

Point-cloud processing

Sensor fusion

ICP

Tools

ROS 2 (rclcpp/rclpy)

TensorRT

ONNX

PCL

Open3D

GTSAM

Job description

About Origin Origin (previously 10xConstruction) is building general-purpose autonomous robots for US construction to tackle rising costs, safety risks, and labor shortages. Our modular, multi-trade platform combines purpose-built hardware with real-time site intelligence to navigate complex environments and execute tasks with precision. Trained in high-fidelity simulation and already deployed on live sites, our robots deliver 5x faster execution, 250%+ margin expansion, and significant cost savings. Join India's most talent-dense robotics team consisting of individuals from IITs, Stanford, UCLA, etc.

About Origin Origin (previously 10xConstruction) is building general-purpose autonomous robots for US construction to tackle rising costs, safety risks, and labor shortages. Our modular, multi-trade platform combines purpose-built hardware with real-time site intelligence to navigate complex environments and execute tasks with precision. Trained in high-fidelity simulation and already deployed on live sites, our robots deliver 5x faster execution, 250%+ margin expansion, and significant cost savings. Join India's most talent-dense robotics team consisting of individuals from IITs, Stanford, UCLA, etc.

About The Role

As a Perception Engineer at Origin (Formerly 10xconstruction), you will help our autonomous drywall-finishing robots "see" the job-site. You'll design and deploy perception pipelines—camera + LiDAR fusion, deep-learning vision models, and point-cloud geometry to give the robot the awareness it needs.

• Develop and deploy 3D perception components for geometric scene understanding using depth sensors, LiDAR, and RGB cameras.
• Process and interpret spatial data (point clouds, depth maps, image streams) for environment modeling and task planning.
• Train and integrate deep-learning models for 3D semantic understanding and quality evaluation.
• Design robust sensor fusion strategies that combine visual, inertial, and spatial data for scene reconstruction and robot localization.
• Benchmark and optimize perception models for deployment on edge compute platforms (e.g., NVIDIA Jetson) using tools like TensorRT or ONNX.
• Collect and curate high-quality datasets (real and synthetic); automate training pipelines and experiment tracking.
• Perform and validate sensor calibration (camera, LiDAR, IMU) and develop checks to maintain calibration quality over time
• Collaborate across robotics teams (manipulation, navigation, cloud) to deliver production-ready perception stacks for autonomous operation in dynamic construction environments

Requirements

• Solid grasp of linear algebra, probability and geometry; coursework or projects in CV or robotics perception
• Proficient in Python 3.x and C++17/20; comfortable with git and CI workflows
• Experience with ROS 2 (rclcpp / rclpy) and custom message / launch setups
• Familiarity with deep-learning vision (PyTorch or TensorFlow)—classification, detection or segmentation
• Hands-on work with point-cloud processing ; know when to apply voxel grids, KD-trees, RANSAC or ICP
• Exposure to sensor calibration algorithms

Good to have:

• Experience with ROS 2
• Familiarity with PCL, Open3D
• Exposure to real-time optimization libraries (Ceres, GTSAM)