Senior Vision Systems Engineer

About the role

Design, optimise and productise high‑performance computer vision and AI pipelines for consumer devices, with a focus on real‑time performance, power efficiency and production robustness across embedded platforms.

Key responsibilities

• Design and implement advanced computer‑vision and image‑processing pipelines optimised for real‑time consumer devices.
• Collaborate with ISP, sensor and tuning teams to optimise image quality for downstream AI and UX performance.
• Develop and deploy ML models for visual recognition, enhancement, tracking or scene understanding, and optimise them for edge deployment.
• Implement performance‑critical algorithms in modern C++ for embedded platforms, targeting latency, power, memory and thermal constraints.
• Integrate inference engines (TFLite, TensorRT, ONNX Runtime, etc.) on target SoCs and work with Android/Linux platform teams.
• Define and track KPIs such as FPS, power usage, memory, startup time and accuracy, and profile performance across CPU/GPU/NPU/DSP.
• Mentor engineers, review architecture proposals and support product bring‑up and mass‑production readiness.

Required profile

• Bachelor's or Master's degree in Computer Science, Electrical Engineering or related field.
• 7‑10+ years of experience in computer vision/image processing and shipping at least one consumer product with embedded vision/AI.
• Strong C++ expertise (C++14/17/20) and performance optimisation skills.
• Extensive experience with OpenCV and ML frameworks such as PyTorch, TensorFlow and ONNX.
• Proven experience deploying and optimising ML models on embedded/edge devices.
• Solid debugging and performance‑profiling abilities on embedded Linux or Android systems.

Required skills

• C++14/17/20
• OpenCV
• PyTorch, TensorFlow, ONNX
• Model optimisation (quantisation, pruning, distillation)
• Embedded Linux and Android platforms
• Performance profiling and power/thermal optimisation
• CUDA, OpenCL, Vulkan, OpenGL ES, SIMD
• Camera calibration and ISP interaction
• Android Camera HAL, Yocto‑based systems
• AR, computational photography, video processing, multi‑camera systems