Senior High-Performance LLM Training Engineer

Description

The company is hiring a Senior High-Performance LLM Training Engineer to make large language model training run more efficiently.

The role centres on performance analysis and optimization for LLM training jobs that run across thousands of GPUs. You will work on the company's LLM software stack in PyTorch and JAX, and help shape hardware roadmaps for future GPUs.

The company builds GPUs, the systems and software on top of them, and optimizes deep learning frameworks. It also supplies GPUs to major systems companies and cloud providers, and builds computing for edge devices such as self-driving cars and autonomous robots.

This is a hybrid role. The base salary range is 184,000 USD to 287,500 USD for Level 4 and 224,000 USD to 356,500 USD for Level 5, set by location, experience, and pay for similar roles.

Responsibilities

• Analyze, profile, and optimize AI training workloads on new hardware and software platforms.
• Find and rank the performance problems across state-of-the-art neural networks on GPUs, then solve them.
• Write production-quality code across several layers of the deep learning platform, from drivers up to DL frameworks.
• Build and maintain the company's entries for the MLPerf Training benchmark suite.
• Port key deep learning training workloads to the company's proprietary processor and system simulators for future architecture studies.
• Build tools that automate workload analysis, optimization, and other key workflows.

Requirements

• A PhD in Computer Science, Electrical Engineering, or Computer Engineering and 5+ years of experience, or an MS (or equivalent experience) and 8+ years of meaningful work experience.
• Strong background in deep learning and neural networks, especially training.
• Deep knowledge of computer architecture and working familiarity with GPU architecture fundamentals.
• Proven experience analyzing and tuning application performance, and modelling processor and system-level performance.
• Programming skills in C++, Python, and CUDA.

Conditions

Hybrid work arrangement.

Base salary range is 184,000 USD to 287,500 USD for Level 4 and 224,000 USD to 356,500 USD for Level 5, depending on location, experience, and pay of similar roles.

Applications are accepted at least until November 1, 2026.

This posting is for an existing vacancy.

The company uses AI tools in its recruiting process.

The company is an equal opportunity employer and does not discriminate on protected characteristics.

Benefits

• Highly competitive salaries and a comprehensive benefits package.