We are seeking a high-caliber Robotics Engineer for a high-velocity development team. In this role, you will be responsible for the full-stack development of the robotics platform, moving seamlessly between low-level hardware integration and high-level ML model deployment. You will build the "brain" and "nervous system" of the robots, ensuring they perceive, think, and move with precision in real-time.
Architecture & Development: Design and implement robust, scalable software modules using ROS 2 (Humble/Iron) in both Python and C++.
ML Deployment: Integrate complex ML model inferences into the robotics pipeline using PyTorch, TensorRT, and JIT to ensure low-latency performance.
Sensor Fusion: Lead the integration of vision systems (Stereo, RGBD), IMUs, and other peripherals into the ROS 2 ecosystem.
Navigation & SLAM: Develop and tune spatial awareness capabilities, leveraging Nav2 and vSLAM frameworks for autonomous operation.
Optimization: Utilize CUDA and cuDNN to optimize compute-heavy tasks on edge devices.
ROS 2 Mastery: Deep, hands-on experience with ROS 2 (not just ROS 1).
Polyglot Programming: Proficiency in Python is required; strong C++ skills are highly preferred for performance-critical components.
Full Stack Robotics: Proven ability to handle everything from middleware communication (e.g., LCM) to high-level application logic.
Hardware Interface: Experience integrating and calibrating vision sensors and IMUs.
ML Ops for Robotics: Familiarity with deploying and optimizing models for real-time robotic environments.
Direct experience with vSLAM (Visual SLAM) algorithms.
Advanced knowledge of ROS 2 Nav2 stack customization.
Experience with high-performance middleware beyond standard DDS, such as LCM.
Proficiency in GPU acceleration using CUDA and cuDNN.
Are you the right fit?
You thrive in high-velocity environments. You prefer shipping code over endless meetings, and you have the "hacker" mindset required to solve unconventional problems, paired with the "engineer" mindset required to build reliable, production-grade systems.