Tech Stack

Description

You will build the "Brain" of Motional's ML-powered multimodal data mining framework, Omnitag. You'll design massive teacher models, distill them into hyper-efficient student models for real-time data discovery, and apply reinforcement learning to optimize data mining workflows, directly accelerating the autonomy stack development.

Requirements

• BS in CS, ML or equivalent with 6+ years ML engineering experience
• Strong experience with model distillation or teacher-student training
• Proven experience with reinforcement learning in production or research
• Expert-level proficiency in Python and ML frameworks (PyTorch, TensorFlow, or JAX)
• Experience deploying ML models in cloud environments and optimizing for inference

Responsibilities

• Architect and train teacher-student model frameworks for multimodal sensor data
• Build RL-based policy learning and reasoning systems for autonomous driving
• Optimize model deployment for real-time inference on GPU/CPU clusters
• Research and integrate agentic workflows for autonomous reasoning
• Drive production reliability and operational excellence