Senior/Staff Software Engineer (Machine Learning & System Optimization)

TL;DR

Senior/Staff Software Engineer (Machine Learning & System Optimization): Orchestrating and allocating system capacity for core perception models on autonomous vehicles with an accent on efficient inference and resource management. Focus on optimizing large-scale models, writing custom CUDA kernels, and implementing low-latency C++ code for real-time execution on edge devices.

Location: Must be based in or able to work from Foster City, CA, Boston, MA, or Seattle, WA (Hybrid)

Company

Zoox is an autonomous vehicle company developing a fully autonomous, purpose-built vehicle and the supporting ecosystem to provide mobility-as-a-service.

What you will do

• Allocate and distribute system resources (CPU/GPU/interconnect) to various models and inference engines running on the robot.
• Spearhead cross-cutting initiatives to improve compute utilization through model sharing, fusion, and advanced scheduling strategies.
• Optimize large-scale models (Multi-Modal Sensor Fusion, LLMs, VLMs) using quantization, pruning, and parameter-efficient fine-tuning.
• Architect and implement model conversion and compilation pipelines using TensorRT for edge deployment.
• Write production-level, low-latency, and memory-safe C++ and CUDA code for real-time inference.

Requirements

• Deep experience in system and performance optimization for CPU/GPU systems.
• Expertise in real-time system constraints including processing latency, memory utilization, and bandwidth pressure.
• Proficiency in model quantization (PTQ, QAT) and mixed-precision inference frameworks.
• Experience developing custom ML OPs and TensorRT Plugins with efficient CUDA kernel implementations.
• Production-level C++ (14/17/20) and Python programming skills.

Nice to have

• Prior experience in high-performance robotics applications such as AV, drones, or robotics.
• Familiarity with SOTA autonomous driving perception algorithms (BEV, 3D Occupancy Networks).
• Experience with end-to-end autonomous driving paradigms and edge deployment technologies like TensorRT-LLM.