Software Engineer, ML Infrastructure

What you'll do

• Develop core data management systems for indexing multi-modal sensor data and managing perception use-cases.
• Build and operate infrastructure for model training and orchestration across cloud and on-premise compute environments.
• Create and maintain internal tools and SDKs to support perception engineers in developing autonomous systems.
• Integrate frontier models into the pipeline for data augmentation, curation, and synthetic data generation.
• Implement semantic search capabilities across broad data holdings to improve discovery and indexing.
• Develop systems for real-time monitoring of training jobs and tracking model evaluation reports.
• Maintain a model registry to facilitate the deployment of improved models to edge devices.
• Collaborate with internal business teams to deliver technical solutions that meet specific project requirements.

What we're looking for

• Build and operate core model training systems and multi-modal sensor data management infrastructure.
• Develop tools and SDKs for perception engineers to manage model training and evaluation.
• Integrate frontier models for data augmentation, curation, and synthetic data generation.
• Develop systems for monitoring the ML lifecycle, including training jobs and model registries.
• Demonstrate strong proficiency in modern Python and exposure to Rust and the Rust/Python toolchain.
• Experience with deep learning frameworks like PyTorch or TensorFlow and distributed training paradigms.
• Experience with MLOps orchestration tools and ML observability platforms.
• U.S. Person status is required; eligibility for a Top Secret clearance is desirable.