RT & Kernel Engineer (human)

We are scaling our real-time embedded Linux capability across NEURA's current and next-generation robot platforms — cobots, mobile manipulators, humanoids, quadrupeds, and the shared Compute Unit. This role owns the operating system, kernel, and board-support layer that every control and communication stack on these platforms depends on.

You will collaborate closely with the RT Communication Bus engineer, hardware, simulation, and safety / certification stakeholders to deliver platforms that are deterministic, maintainable, and suitable for a controlled product lifecycle.

• Real-time Linux platforms — both co-kernel and kernel-preemption approaches — including RT scheduling policy and interrupt-latency budgeting to meet the timing requirements of the control and estimation loops running on the platform (kHz-class whole-body control and high-rate state estimation).

• SOM board bring-up and the Yocto BSP for each robot platform hardware revision: boot flow, kernel configuration, device trees, and driver work at the C/C++ boundary, with hands-on debugging on real hardware.

• Yocto layers, recipes, and images for development, validation, and release; structured evaluations and migrations across multiple SoCs and board variants as hardware evolves.

• WCET analysis across RT loop configurations, and RT memory management practice (memory locking, huge pages, NUMA-aware allocation, priority inversion avoidance).

• Co-review of WCET budgets and cycle timing with the RT Communication Bus engineer at the OS / fieldbus boundary, ensuring scheduling, interrupt, and synchronisation assumptions hold under load.

• Safety integration at the kernel level (e.g. E-stop chain wiring, ISO 10218 safety PLC integration); design participation, documentation, and traceability for controlled change practices.

• Build, flash, and test automation; technical documentation kept current for the wider engineering organisation.
  

• Embedded Linux BSP engineering background with hands-on Yocto / OpenEmbedded experience on production hardware.

• Hands-on real-time Linux kernel patching and tuning experience, with practical scheduling-oriented reasoning (SCHED_FIFO/RR, priorities, affinity, interrupt path).

• Working knowledge of WCET analysis methodology and tooling (e.g. OTAWA, Chronos, or vendor equivalents).

• Solid C and C++ at the kernel/driver boundary; device tree authoring; practical scripting for engineering workflows.

• Experience on ARM SoCs with multi-cluster configurations (e.g. Cortex-A + Cortex-R), or comparable platforms.

• A collaborative working style: shared design, constructive code review, proactive communication, and reliable coordination across hardware, communications, and safety disciplines. Strong teamwork is essential for this role.

• Bachelor's degree in a relevant field, or equivalent practical experience.

Nice to Have

• EtherCAT familiarity sufficient for OS / fieldbus-boundary co-review (WCET budgets and cycle timing).

• Exposure to functional safety engineering practices (ISO 26262, IEC 61508, or ISO 10218) — documentation, evidence, review cycles. Certification sign-off is not required for this role.

• Experience with PTP / IEEE 1588 hardware timestamping at the kernel network driver level.

• Background in robotics, motion control, or industrial automation with strict timing expectations.

• Interest in growing scope as the platform team expands.

• Become part of an agile company, actively shape topics and benefit from flat hierarchies in a highly motivated team.

• Enjoy an attractive salary, flexible working hours and 30 days of vacation.

• The freedom to contribute your own ideas and drive them forward.

• Celebrate successes together with company events.

• Take advantage of our corporate benefits program.

• And even more fun with great colleagues.