About Origin Origin (previously 10xConstruction) is building general-purpose autonomous robots for US construction to tackle rising costs, safety risks, and labour 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 The Embedded Vision Engineer owns the robot's imaging system end-to-end — from camera and lens selection through driver bring-up and ISP tuning to delivering clean, timestamped frames to the perception stack.
You'll work across hardware, electrical, and perception teams, translating their requirements into camera subsystem specs and building the NVIDIA Jetson native acquisition pipeline that feeds every downstream algorithm on the robot.
Key Responsibilities Select and qualify camera hardware (sensors, lens assemblies, camera modules) against requirements from perception, navigation, and manipulation teams — making trade-off decisions on resolution, shutter type, dynamic range, FOV, focal length, and spectral response for each use case on the robot.
Bring up and maintain camera drivers on NVIDIA Jetson (Orin, Orin Nano, Thor) for GMSL2, MIPI CSI-2, and GigE Vision (PoE) interfaces — writing and patching V4L2 drivers, device-tree overlays, and NVIDIA-specific components (libargus, nvarguscamerasrc, nv_multimedia_api).
Own optics and imaging quality — specify lens parameters, tune ISP settings (white balance, exposure, gain, HDR), run intrinsic/extrinsic calibration for multi-camera rigs, and characterise camera performance under construction-site conditions (dust, vibration, spray mist, variable lighting).
Design and implement hardware triggering (PWM/GPIO) and PTP (IEEE 1588) synchronisation to align frame capture across multiple cameras and with other onboard sensors, ensuring sub-millisecond cross-sensor timing.
Build the image-acquisition pipeline (GStreamer / NVIDIA DeepStream) that moves frames from sensor to GPU memory with zero-copy (DMA-BUF, NVMM) handoffs, and specify the illumination system (structured light, LED strobe) synchronised with camera exposure windows.
Required Qualifications and Skills 3–5 years of hands-on work with embedded camera systems — sensor evaluation, camera driver bring-up, and imaging pipeline development on NVIDIA Jetson platforms (Xavier, Orin, or Orin Nano).
Driver-level experience with GMSL2, MIPI CSI-2, or GigE Vision camera interfaces, including device-tree overlays, V4L2 subsystem, and SERDES debugging.
Working knowledge of optics and imaging fundamentals — lens selection (focal length, FOV, aperture, distortion), exposure control, ISP tuning, and camera calibration.
Fluency in C and C++ with comfort reading and patching Linux kernel and V4L2 code; experience building low-latency pipelines with GStreamer, NVIDIA multimedia API, or DeepStream.
Preferred Experiences Hands-on with GMSL2 SERDES ICs (MAX 96712 / MAX 9295) — schematic review, register configuration, link-lock debugging.
Experience with industrial machine-vision cameras (GigE Vision / GenICam ecosystem — FRAMOS, Basler, FLIR, Lucid) or active illumination systems (structured light, LED strobe timing).
Hardware triggering and multi-sensor synchronisation using PTP (IEEE 1588) or GPIO/PWM trigger design; familiarity with ROS 2 camera drivers and robotics middleware integration.
