Senior/Staff Software Engineer - Machine Learning & System Optimization

The role

You will allocate system capacity (CPU, GPU, interconnect) across core perception models running on the vehicle. You will also optimize large-scale models such as LLMs, VLMs and foundation models for real-time inference on edge devices.

What you will do

• Allocate and distribute CPU, GPU and interconnect resources to models and inference engines on the robot
• Lead cross-cutting work on model sharing, fusion and scheduling to improve compute utilization
• Optimize large multi-modal models using quantization (PTQ, QAT), pruning, mixed-precision inference and parameter-efficient fine-tuning (LoRA, QLoRA)
• Build model conversion and compilation pipelines with TensorRT for edge deployment
• Write production-level, low-latency, memory-safe C++ and CUDA code for real-time inference on vehicle systems
• Write custom CUDA kernels and build highly concurrent inference code for real-time, deterministic execution on edge devices

Required qualifications

• Deep experience in system and performance optimization of CPU/GPU systems for low latency or high throughput
• Deep expertise in real-time systems and constraints such as processing latency, memory utilization and memory bandwidth
• Deep expertise in model quantization (PTQ, QAT) and mixed-precision inference (INT8, FP8, FP4, BF16/FP16)
• Proficiency in low-level programming for AI accelerators, including custom ML ops and TensorRT plugins with CUDA kernels
• Production-level C++ (14/17/20) and Python programming, including concurrent, memory-safe real-time inference code for edge devices
• Hands-on experience compressing, accelerating and deploying complex models (LLMs, VLMs, foundation models) on power- and thermal-constrained vehicle SoCs

Nice to have

• Prior experience in high-performance robotics applications such as autonomous vehicles, drones or robots
• Familiarity with autonomous driving perception algorithms (temporal 3D object detection, BEV, 3D occupancy networks) and multi-modal sensor processing (vision, LiDAR, radar)
• Experience with end-to-end autonomous driving paradigms (VLM/VLA models, foundation models) and edge deployment technologies such as TensorRT-LLM

Before you apply

• You do not need to match every listed expectation to apply

About the company

Zoox is developing the first ground-up, fully autonomous vehicle fleet and the supporting ecosystem to bring it to market. It aims to provide next-generation mobility-as-a-service in urban environments.