Live opening · Posted 1 day ago
At a glance
The key details from the original listing.
Your early-applicant advantage
Live timing from JobBeeper.
About the role
Description supplied by the original job listing.
About the Role
You will lead the development of an edge-AI camera platform. It starts with prototyping and model training on NVIDIA Jetson, then moves to production hardware built on Raspberry Pi 4/5 with the Hailo-8L accelerator, or on a custom Jetson-based board. You will own the work from camera bring-up and model deployment through to custom PCB design and production readiness.
Key Responsibilities
• Bring up and tune CSI camera modules (IMX708, IMX477, IMX219 class), including the ISP, libcamera and GStreamer pipelines.
• Train, fine-tune and validate YOLO object detection models using datasets from the real deployment camera.
• Build depth estimation (monocular and stereo) and integrate it with detection.
• Deploy models on Jetson with TensorRT, and on Raspberry Pi with Hailo-8L through ONNX export, Hailo Dataflow Compiler (INT8 calibration) and HailoRT.
• Run the Jetson versus RPi + Hailo feasibility study, covering performance, cost, thermal and supply-chain tradeoffs.
• Lead schematic and PCB design for the custom board, including MIPI-CSI, PCIe and power design.
• Define test plans for accuracy, latency, FPS and thermal behavior, and verify outputs before production release.
• Document the design and mentor junior engineers.
Required Skills
• B.E./B.Tech/M.Tech in Electronics, Electrical, Computer Science or a related field.
• Strong Python and C/C++ on embedded Linux.
• Hands-on experience with NVIDIA Jetson (JetPack, CUDA, TensorRT) and Raspberry Pi.
• Practical computer vision and deep learning experience, including YOLO training and deployment.
• Camera interfacing: MIPI-CSI, V4L2, libcamera, GStreamer.
• Model optimization: ONNX, quantization, edge accelerators.
• Schematic and PCB design in Altium, KiCad or OrCAD, with high-speed signal awareness.
Preferred Skills
• Experience with Hailo, Coral or similar NPU toolchains.
• Stereo vision, calibration, or monocular depth models (MiDaS class).
• Yocto/Buildroot, custom BSP or device-tree work.
• Experience taking a hardware product from prototype to manufacturing (DFM, EMI/EMC, certification).
• Automotive or industrial vision background, including functional safety awareness.
What We Offer
Ownership of a product from first prototype to production, with a competitive salary and growth opportunities.
Work arrangement
No
More openings worth a look
Recently tracked roles with full details and direct application links.