Machine Learning Engineer

Description:

We are looking for a creative & driven Machine Learning Engineer to join our autonomous vehicle team. You will be at the center of our efforts to build intelligent systems that can understand, predict, and safely navigate a complex and dynamic world. You will design and train the models that anticipate what other road users will do next, and turn those predictions into driving decisions. This means working with petabytes of real driving data, building evaluation frameworks that actually correlate with on-road safety, and shipping models that run under hard latency budgets on embedded hardware.

In this production engineering role, you will spend meaningful time on failure analysis, long-tail scenarios, and the gap between offline metrics and on-road behavior. If you are passionate about applying cutting-edge ML to solve high-stakes robotics challenges, we want to hear from you.

What You’ll Do
 

• Design, train, and deploy models for behavioral prediction and motion planning that run on vehicles in real traffic
• Model multi-agent interaction and temporal dynamics — how a merge, an unprotected left, or an occluded pedestrian actually unfolds
• Own the metrics: build evaluation frameworks that correlate with real on-road safety and performance, not just offline loss
• Diagnose long-tail failures from real driving logs and close the loop back into training data and model design
• Optimize trained models for real-time inference under strict latency, memory, and compute constraints on embedded hardware
• Build and maintain data pipelines that process, clean, and label large-scale vehicle sensor and simulation datasets
   

What You’ll Need

Domain experience (required):
 

• Hands-on experience with at least one of: behavioral or trajectory prediction, motion planning, decision-making under uncertainty, or closely adjacent autonomy work (navigation, SLAM, control, or perception-for-planning) for autonomous vehicles, mobile robots, drones, or comparable physical systems
• Experience deploying machine learning to real hardware operating in the physical world, under real-time or resource constraints. Simulation-only or offline-only experience does not meet this bar.
   

Engineering (required)
 

• Strong Python and production experience with a modern deep learning framework (PyTorch, TensorFlow, or JAX)
• Proficiency in C++ (or Rust) for performance-critical inference and integration code
• Demonstrated ownership of a system from prototype through deployment, including debugging it after it shipped