Jobs at Mobileye

Analyze autonomous driving performance from real-world testing, identify root causes of system issues, translate findings into actionable insights for engineering, collaborate with cross-functional teams, and provide high-impact feedback to improve Mobileye's AV stack and user experience.

Own the sim-to-real bridge for a full-size humanoid: build and maintain digital twins from CAD, model actuators from measured data, run system-identification campaigns, design domain randomization in Isaac Sim, manage deployment pipelines, generate synthetic data, and quantify and reduce the sim-to-real gap.

Develop and deploy generative and computer vision models (esp. diffusion-based) for autonomous driving. Own model training and adaptation, implement scalable Python/PyTorch solutions, collaborate with perception teams to integrate synthetic data, and drive research from prototypes to real-world AV use cases.

Design, build, and scale hybrid (AWS + on-prem) infrastructure for a large data platform supporting DL/ML. Lead projects from architecture to rollout, improve availability, resiliency, performance, and cost efficiency, develop shared templates and CI/CD tooling, evaluate new technologies, and act as a technical owner for observability and infrastructure standards.

Design and own compact, high-current motor controller PCBs for humanoid robot joints from schematic to bring-up, including inverter stage, sensing, protections, encoder and sensor interfaces, MCU/communication PHY integration, high-density thermal layout, EMI/EMC support, and fast iteration with suppliers.

Define Linux-based OS solutions for PCIe connectivity between peripherals and multiple EyeQ SoCs; design and implement the PCIe SW stack for cross-chip neural networks; collaborate on HW/SW co-development with SoC architects; perform performance analysis, feasibility studies, PoCs and optimizations; guide and mentor Linux kernel development engineers.

Develop and implement perception, mapping, computer vision, and 3D/sensor-fusion algorithms for AV/ADAS using Python and C++; work with image-based data, solve complex algorithmic problems, and collaborate with engineering and research teams to deliver real-world solutions.

Design and architect DFT solutions across SoCs (MBIST, SCAN, ATPG, LBIST), develop RTL, verify in gate-level simulation, ensure ATPG meets ISO26262 safety, analyze coverage/yield, and support production and silicon debug.

Develop and optimize algorithms for continuously updating large‑scale 3D HD map creation. Implement and maintain production Python code, analyze large datasets using scientific and big‑data tools, and take research prototypes through production deployment to improve road safety.

Perform pre-silicon system-level verification of automotive AI accelerators: define test plans, develop and run tests in simulation/emulation, build test environments and verification collateral, perform fullchip/SoC verification, debug with EDA tools, and validate IP integrations and industry-standard protocols.

Maintain and evolve large-scale distributed infrastructure across Kubernetes, on-prem, and cloud; design and automate Infrastructure as Code, GitOps, and CI/CD tooling; troubleshoot Linux, networking, storage, and CI/CD issues; collaborate cross-functionally to improve developer experience and platform reliability.

Design, train, optimize, and evaluate production-grade deep learning perception models (detection, tracking, segmentation) for autonomous vehicles using large driving datasets; analyze edge cases and implement scalable training and validation pipelines.

Research, design, implement and optimize lidar-based computer vision algorithms in Python and C++; evaluate and improve models on large-scale real-world vehicle data; collaborate cross-functionally to meet ADAS and autonomous driving requirements.

Architect and implement embedded Linux userspace components and startup/service management for automotive SoCs. Develop FOTA/update frameworks, low-level C/C++ daemons, optimize boot/resource usage, perform system-level root-cause analysis, and support board bring-up for EyeQ platforms.

Design and deliver high torque-density humanoid joint actuator modules end-to-end: motor selection, gear reduction, thermal and mass budgets, sensing integration, packaging, structural design, prototype fabrication, and joint-level dyno and life testing while collaborating with electrical and control teams.

Develop and optimize a deep learning compiler stack to map neural networks (e.g., PyTorch) onto custom Mobileye hardware. Implement and optimize neural-network operations, use MLIR/LLVM, collaborate with open-source community, and influence future hardware/software design for efficient in-vehicle inference.

Build automation tools and AI-powered internal solutions to improve engineering productivity and mass production of autonomous driving systems. Diagnose complex issues, design workflow orchestration and large-scale deployment tooling, and collaborate with R&D, validation, testing, and global teams to drive reliable production delivery.

Owner of block-level physical design from RTL to GDS: floorplanning, synthesis, STA, netlist preparation, place-and-route, and signoff verification while optimizing power, performance, and area.

Design, train, and iterate whole-body reinforcement-learning policies in Isaac Sim/Newton for walking, balance recovery, manipulation, and loco-manipulation. Own reward design, curricula, large-scale training infrastructure, sim-to-real transfer, and on-hardware deployment in collaboration with Sim2Real and platform engineers.

Lead pre-silicon system-level verification of automotive AI accelerators and SoCs: define test plans, develop and run tests in simulation/emulation, build verification environments, and support full-chip verification through delivery.