AI Systems Engineer (OCI/AI Infrastructure)

Oracle•Nashville, TN

About The Position

Oracle Hardware Platform Development Engineering is seeking an AI Systems Engineer to evaluate and characterize next-generation GPU and AI accelerator platforms for Oracle Cloud Infrastructure (OCI). This hands-on engineering role brings up new hardware platforms, enables AI training and inference software stacks, runs representative workloads, and analyzes system performance under real operating conditions. The engineer investigates whether workloads are limited by HBM or memory bandwidth, compute, scale-up, or scale-out, and characterizes power, thermals, memory behavior, utilization, scaling, and performance efficiency. Working directly in the lab, the engineer debugs hardware/software integration issues, designs and executes experiments, and develops data-driven insights into system behavior beyond benchmark results. The role includes comparative architecture analysis across GPUs and emerging AI accelerators, evaluating architectural tradeoffs, and translating findings into recommendations about platforms suited to specific AI training and inference workloads. The engineer works with internal hardware and software teams and technology partners to help shape Oracle’s next generation of high-performance AI infrastructure, supporting the training and deployment of next-generation LLMs and generative AI models.

Requirements

Solid knowledge of AI, GPU, and/or AI/CPU platform architectures and their capabilities.
Experience with the architecture, design, and implementation of modern server platforms involving multiple architectures and vendors, including x86 and ARM.
Strong communication skills, including the ability to explain complex technical issues across engineering disciplines and succinctly communicate issues to executives.
Experience with the latest high-speed buses and interconnects used in modern compute and AI platforms, including their startup, connectivity, operational robustness, and performance metrics.
Strong debugging skills and experience analyzing and triaging AI applications from the application level to the hardware level.
Bachelor’s degree or higher in Computer Science or a related technical field, or equivalent experience.

Nice To Haves

At least 8 years of experience developing software infrastructure for large-scale AI systems.
Hands-on experience maintaining or building ML training stacks involving CUDA, ROCm, NCCL, XLA, or similar technologies.
Experience benchmarking AI workloads across different architectures.
Background working with large-scale clusters.
Good understanding of the internals of deep learning frameworks such as PyTorch, TensorFlow, JAX, and Ray.

Responsibilities

Bring up new hardware platforms and enable AI training and inference software stacks.
Run representative workloads and analyze system performance under real operating conditions.
Determine whether workloads are limited by HBM or memory bandwidth, compute, scale-up, or scale-out.
Characterize power, thermals, memory behavior, utilization, scaling, and performance efficiency.
Debug hardware/software integration issues in the lab.
Design and execute experiments and develop data-driven insights into system behavior.
Compare GPU and emerging AI accelerator architectures, evaluate tradeoffs, and recommend platforms for specific AI training and inference workloads.
Troubleshoot complex hardware/software interaction issues, including vLLM compilation failures on ROCm, CUDA memory leaks, distributed runtime failures, and kernel-level inconsistencies.
Profile compilation graphs, training workloads, and runtime execution to optimize performance and eliminate bottlenecks.
Work with internal hardware and software teams and technology partners to help shape future AI infrastructure.

Benefits

Medical, dental, and vision insurance, including expert medical opinion.
Short-term and long-term disability insurance.
Life insurance and AD&D.
Supplemental life insurance for employees, spouses, and children.
Health care and dependent care Flexible Spending Accounts.
Pre-tax commuter and parking benefits.
401(k) Savings and Investment Plan with company match.
Paid time off, including flexible or accrued vacation depending on eligibility and position; vacation accrual is described for employees working at least 20 hours per week.
11 paid holidays.
Paid sick leave: 72 hours upon hire, refreshed annually, with unused time carrying over up to a 112-hour cap.
Paid parental leave.
Adoption assistance.
Employee Stock Purchase Plan.
Financial planning and group legal benefits.
Voluntary auto, homeowner, and pet insurance.
May be eligible for bonus, equity, and compensation deferral.