ML Engineer Intern: Efficient Vision Models for Autonomy

Waymo LLC; Mountain View (CA); Hybrid; USD 96,000 - 117,000; Full time

Benefits offered by this job

Housing/relocation bonus

Medical, dental, and vision insurance

Free meals onsite

Free shuttle access

Job summary

Waymo LLC in Mountain View is offering a hybrid onsite internship with a focus on model optimization for perception and foundation models used in autonomous driving. You will work on optimizing and deploying ML models, collaborating with cross-functional teams across hardware and software domains.

The role emphasizes Python and DL frameworks (JAX, Flax, PyTorch, or TensorFlow) and includes hands-on tasks in quantization, latency reduction, and multi-task learning for real-time on-vehicle

Qualifications

• Pursuing a PhD or Master’s in CS, EE, ML, Robotics, or related field.
• Strong software and DL development skills in Python and ML frameworks.
• Solid understanding of foundation models and multi-task learning.
• Experience with model compression, LoRA/QLoRA, or quantization techniques.

Responsibilities

• Design, implement, and benchmark LoRA/QLoRA modules in JAX/Flax for Vision Transformer backbones.
• Develop dual-level distillation pipelines to mitigate multi-task regressions during large-scale data scaling.
• Collaborate with model optimization, quantization, and latency teams to validate static weight folding and low-precision quantization.
• Conduct extensive empirical ablations and evaluate perception metrics on large-scale autonomous driving datasets across geographic domains.

Skills

Python

JAX/Flax

PyTorch/TensorFlow

Transformer architectures

Multi-task learning

Education

PhD or Master’s in CS/EE/ML/Robotics

Tools

JAX

Flax

TensorFlow

PyTorch

Job description

Waymo LLC in Mountain View is offering a hybrid onsite internship with a focus on model optimization for perception and foundation models used in autonomous driving. You will work on optimizing and deploying ML models, collaborating with cross-functional teams across hardware and software domains.

The role emphasizes Python and DL frameworks (JAX, Flax, PyTorch, or TensorFlow) and includes hands-on tasks in quantization, latency reduction, and multi-task learning for real-time on-vehicle