Founding AI Research Engineer - Robot Learning
Core
Designing and deploying learned policies for autonomous robots performing interior construction tasks in unstructured environments.
Role type
Founding AI Research Engineer (Robot Learning)
Builds
Modular robots for interior construction using Multi Agent Action Expert architecture
Domain
Construction Robotics / Physical AI
Deliverable
production ML models
Required skills
Imitation learning, Reinforcement learning, Vision-language models, Robot learning from demonstration, Sim-to-real transfer, Edge deployment optimization, ROS2 middleware, Simulation systems (Isaac Sim), GPU profiling (TensorRT, CUDA)
Preferred skills
VLA architectures (π0, OpenVLA, RT-2, Octo), Foundation model fine-tuning, Teleoperation data pipelines, World model architectures (DreamerV3, V-JEPA), Construction domain experience
Technologies
PyTorch, Python, Jetson AGX Orin, TensorRT, ONNX, CUDA, ROS2, Isaac Sim, GELLO, SpaceMouse, VR
Responsibilities
Train and deploy VLA models for contact-rich manipulation; Build data flywheels including teleoperation pipelines and online correction; Research world models for surface state prediction and anomaly detection; Optimize models for edge latency and memory constraints; Design interfaces between learned policies and deterministic safety layers
Seniority
Founding, hands-on IC
