Staff Software Engineer: Deep Learning Acceleration & Performance

AURORA; Pittsburgh (Allegheny County); Hybrid; USD 171,000 - 247,000; Full time

Benefits offered by this job

Annual bonus

Equity compensation

Benefits

Job summary

Aurora is seeking a Staff Software Engineer focused on Deep Learning Acceleration to optimize DL networks for autonomous vehicles. You will analyze performance, reduce latency, and tune GPU workloads across onboard and data-center training environments.

Join a fast-paced team advancing self-driving tech with cross-functional collaboration, hybrid in-office work, and ongoing opportunities in research and deployment.

Qualifications

• Minimum 5+ years of professional software engineering experience.
• BS, MS, or PhD in Computer Science or a related field.
• Strong programming skills in CUDA, C++ and Python.
• Extensive experience in HPC and parallel programming to optimize GPU usage and latency.
• Proficiency with performance analysis tools (Nsight Systems/Compute) and roofline techniques.
• Hands-on DL/ML optimization at framework level (PyTorch or TensorFlow).
• Solid understanding of computer vision fundamentals and transformer-based DL architectures.
• Excellent analytical and troubleshooting skills for large codebases.
• Able to learn new tools quickly and work across multidisciplinary teams.
• Comfortable in Linux/Unix environments.

Responsibilities

• Conduct performance analysis and optimization of Deep Learning networks running on the AV.
• Optimize software architecture, system performance, and latency for DL workloads.
• Deploy deep learning models on the AV and train in data centers.
• Troubleshoot performance issues using profiling and roofline techniques.
• Collaborate with cross-functional teams to improve efficiency of self-driving tech.

Skills

CUDA

C++

Python

High-performance computing

Roofline model

Performance profiling

DL/ML frameworks

PyTorch

TensorFlow

Computer vision basics

Transformer networks

Linux/Unix

Communication skills

Education

BS/MS/PhD in Computer Science or related field

Tools

NVIDIA Nsight Systems

NVIDIA Nsight Compute

TensorRT

OpenAI Triton

Mojo

Job description

Aurora is seeking a Staff Software Engineer focused on Deep Learning Acceleration to optimize DL networks for autonomous vehicles. You will analyze performance, reduce latency, and tune GPU workloads across onboard and data-center training environments.

Join a fast-paced team advancing self-driving tech with cross-functional collaboration, hybrid in-office work, and ongoing opportunities in research and deployment.