Atoms
Staff Machine Learning Engineer
$273K–$345K
San Francisco, CA +1 · Staff · Full-time · Posted Jun 26, 2026
- Location
- San Francisco, CA
San Francisco, California, United States - Experience
- 10+ years of non-internship professional MLE experience.
About the job
Develop machine-learning systems for autonomous transport, spanning world-model research, model optimization for vehicle hardware, and mining long-tail driving scenarios from sensor data.
What you'll do
- Research and develop reinforcement learning, distillation, and multimodal models for trajectory planning and vehicle actuation; build interactive world models from sensor logs.
- Optimize, profile, and deploy models on vehicle edge hardware, including through hardware-accelerated runtimes.
- Build data pipelines to identify rare driving scenarios and curate data for training and simulation testing.
What they ask for
- 10+ years of non-internship professional MLE experience.
- Expertise in AI Transformers for robotics, physical actuation, or spatial-temporal data, and experience designing or training multimodal, VLA, or generative Diffusion models.
- Experience with model optimization, edge deployment, or large-scale physical-dataset curation, depending on specialization.
- Programming proficiency in Python and experience with PyTorch or JAX; C++ is also requested for some specializations.
Nice to have
- Experience with multi-task learning, Birds-Eye-View (BEV) frameworks, representation learning, or data tokenization.
- Familiarity with low-level camera/sensor interfaces and robotics hardware.
Pay
$273,000 - $345,000 per year (base salary)
Tools
PyTorch, JAX, Python, C++, TensorRT, CUDA, Cameras, LiDAR
AI focus
Develop physical AI for autonomous transport, including multimodal perception-to-actuation models, world models, reinforcement learning, post-training optimization, and data curation for rare driving scenarios.
Model-written summary · the original posting is authoritative