SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Apptronik is a Series B human-centered robotics company developing AI-powered humanoid robots (Apollo) to support humanity across manufacturing, logistics, healthcare, and beyond. As Principal Robotics Machine Learning Engineer, you will be a core technical contributor responsible for the end-to-end lifecycle of learned manipulation—from data engine design through training to on-robot deployment.
You will implement and deploy state-of-the-art ML models and algorithms to achieve world-class performance on open-world object manipulation tasks with physical hardware. Key responsibilities include driving the data engine for manipulation learning (collection strategy, curation, annotation, synthetic/real data mix) in partnership with data infrastructure teams; managing the entire development cycle from simulation prototyping to robust model transfer and fine-tuning on the robot; identifying and prioritizing performance bottlenecks by distinguishing between classical vision issues (calibration, geometry, sensor fusion) versus model/data optimization; and acting as a force multiplier across the organization through code reviews, technical rigor, and mentoring.
You must have 5+ years shipping ML models on robotic systems in production, expertise in CNNs, transformers, and VLAs, proficiency in Python, C++, and PyTorch for real-time robotic software stacks, experience with data pipelines (collection, annotation, synthetic data), and deep knowledge of computer vision (6D pose estimation, camera calibration, point cloud processing, visual servoing). Hardware bring-up experience with high-DOF manipulators is essential. Good-to-have skills include MLOps, VR/haptic teleoperation, physics simulation (IsaacSim, MuJoCo, Drake), and logistics automation background.
Education: BS/MS/PhD in Robotics, Computer Science, Electrical Engineering, or related field. Experience: 7+ years relevant (or 5+ with PhD) in robotic manipulation or complex motion control, with proven track record deploying complex algorithms on physical hardware. The role operates at startup pace—intense and focused but sustainable—with high ownership, fast feedback, and low bureaucracy.
