Luma AIPalo Alto, CADevelops and optimizes inference engines for multimodal AI models, integrating new architectures, building scheduling systems, and managing large-scale GPU deployments. Requires strong Python, model serving frameworks like PyTorch/vLLM, and Kubernetes expertise.
Must have:
Bonus points:
Must have:
Nice to have:
The base pay range for this role is $187,500 – $395,000 per year.
Builds foundational ML platform infrastructure including model serving pipelines, GPU scheduling systems, and CI/CD for large-scale multimodal AI models. Requires 5+ years in distributed systems with expertise in Python, Kubernetes, and AWS.
Builds and optimizes distributed training infrastructure for large-scale multimodal AI models across thousands of GPUs. Requires deep expertise in PyTorch, CUDA, parallelization techniques, and GPU clusters.
Build and operate Chime's ML platform on AWS, including distributed training systems, feature stores, data pipelines, and CI/CD tooling. Partner with ML teams to improve reliability, observability, and developer experience for production models.
Builds production-ready agentic AI systems including runtimes, orchestration, reliability, observability, and integrations with LLMs/APIs. Requires strong backend experience, shipped agent/LLM systems, and production reliability expertise.
As a Software Engineer on the Behavior Capabilities team, you will develop and implement algorithmic advancements to expand the robot's driving abilities in complex scenarios, focusing on improving trip progress and vehicle uptime.