Deep Learning Performance Architect - New College Grad 2027

Description

The company is at the front of generative AI. The team is hiring for the 2027 new college graduate program and works on GPU compute architecture and performance engineering, covering pre-silicon modeling, data-driven architecture exploration, and GPU microarchitecture improvements for LLM inference, agentic AI, and other generative AI workloads.

In this role you will help make large language models, agentic AI systems, and other generative AI models run more efficiently and scale better on the company's platforms. You will learn pre-silicon performance modeling, correlation, bottleneck analysis, architecture studies, and system-level analysis while working alongside kernel, hardware, and performance engineers.

The work feeds studies that shape how next-generation LLMs, agentic AI, and generative AI models perform in data center and edge settings. You will help build system-level models and correlation methods that produce accurate projections and point the way to efficiency gains.

Responsibilities

• Build or improve pre-silicon and system-level performance models to understand workload behavior, architecture tradeoffs, and projection accuracy.
• Develop correlation methods that compare model projections with lower-level analyses or reference data, so the models get more faithful.
• Study how microarchitecture and system design choices shape end-to-end performance, efficiency, scalability, and projection fidelity.
• Find bottlenecks in model execution, kernels, data movement, and overall system behavior.
• Explore and assess architecture ideas that raise performance, efficiency, or model fidelity.
• Work with kernel, hardware, and framework teams to understand end-to-end behavior for LLMs, agentic AI systems, and other generative AI models.
• Use profiling and benchmarking tools to support model development, analysis, and correlation.
• Present your findings through clear analysis and visualizations.

Requirements

• Hold a BS, MS, or PhD in computer science, computer engineering, electrical engineering, or a related field, or have equivalent experience.
• Program well in C++ and Python, with solid software engineering fundamentals.
• Have a basic understanding of modern deep learning models such as Transformers and LLMs.
• Show interest in pre-silicon modeling, GPU architecture, correlation, system-level behavior, or systems optimization.
• Have exposure to computer architecture, parallel programming, performance analysis, correlation, system modeling, or related systems topics through coursework, research, internships, or projects.
• Be familiar with at least one ML framework, modeling environment, profiling tool, or efficiency-focused software stack, such as PyTorch, TensorRT-LLM, Triton, Nsight, vLLM, or similar.
• Reason clearly about performance tradeoffs and debug technical problems.
• Communicate well and work effectively with a team.

Conditions

Helps you stand out: experience with CUDA programming, GPU performance analysis, performance modeling, correlation, system-level analysis, or profiling tools.

Helps you stand out: hands-on work with performance-sensitive systems, computer architecture, or ML infrastructure through coursework, projects, internships, or research.

Helps you stand out: curiosity about how systems work internally and a strong drive to solve hard problems.

Helps you stand out: a habit of learning quickly, taking ownership, and making steady progress on open-ended work.

Helps you stand out: ability to work independently, explore unfamiliar technical areas, and follow through on personal, academic, or open-source projects in GPU architecture, performance engineering, ML systems, agentic AI systems, or developer tools.

Base salary depends on location, experience, and what similar employees are paid: 124,000 USD to 195,500 USD for Level 2, and 152,000 USD to 241,500 USD for Level 3.

Applications are accepted at least until October 31, 2026. This is 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.