Define and implement robust ROS architectures including workspace structure, package management, and node communication paradigms.
Configure ROS QoS policies and security features, runtime management (Executor), and modify/add client libraries.
Develop, maintain, and containerize (Docker) complex launch configurations and parameter management systems.
Design, code, test, and debug software nodes, services, and actions in C++ or Python to control Physical AI systems.
Customize Linux kernel performance, scheduling, and resource allocation to ensure low-latency performance for robotics applications and other safety-critical tasks.
Identify and resolve low-level system failures, debugging race conditions, and handling memory management to improve stability.
Design and manage the communication framework (e.g., DDS, MQTT) that allows nodes, sensors, and actuators to exchange data in real-time.
Write and optimize device drivers for sensors (Lidar, cameras, IMUs) and actuators (motors, servo controllers) to work with Linux and ROS.
Work closely with hardware engineers, application teams, and QA testers, and create technical documentation for system architecture.

Masters/PhD in Computer Science, Machine Learning, AI, Robotics, or related field.
Masters (10 – 15 years)/ PhD (3-5 years) of experience building robotics/Physical AI systems.
Strong understanding of operating system concepts, including process management, multi-threading, concurrency, memory management (paging/virtual memory), and interrupt handling.
Expert knowledge of Ubuntu Linux, terminal usage, shell scripting (Bash), and networking (TCP/IP).
Deep knowledge of ROS1 & ROS 2, especially ROS 2 and its middleware components (DDS, FastDDS/CycloneDDS).
Experience with Behavior Trees and other robot behavior orchestration frameworks.
Deep proficiency in C and C++ is essential for kernel-level development, with Rust increasingly required for modern systems, and Assembly for debugging and performance tuning.

Opportunity to work on state-of-the-art embodied AI models powering real robots.
Combination of research and deployment — not just writing models, but seeing them act in the physical world.
High-impact work on cutting-edge robotic autonomy and swarm behaviours.
Exposure to both AI and hardware-level execution environments.

Principal Physical AI Engineer – ROS Architect

Key skills