Intel Hiring AI System Performance Engineer (Bangalore)

Intel Corporation; Bengaluru; Hybrid; INR 3,000,000 - 4,000,000; Full time

Job summary

Intel Corporation in Bengaluru is seeking an AI System Performance Engineer to analyze training and inference workloads, build analytical models, and identify bottlenecks affecting performance on Intel architectures. You will explore AI workloads, GPU interconnects, memory hierarchies, and deployment scenarios to guide configuration choices.

The role requires a hands-on, self-driven approach and the ability to quickly learn new workloads, tools, and architectures, with Intel's hybrid work model

Responsibilities

• Enable new AI workloads in analytical simulators, and model emerging multi-GPU deployment, parallelization, and communication strategies as simulator capabilities.
• Build analytical models of GPU architectures, covering core, memory hierarchy, and training/inference deployments; produce performance estimates to reveal bottlenecks.
• Integrate and modify open-source simulators or analytical models for Intel-specific architecture.
• Perform what-if analyses over core count, memory, bandwidth, interconnect, and deployment options to deliver optimization recommendations.

Job description

About the Role

Intel is looking for a talented and motivated AI System Performance Engineer. The role builds a strong understanding of AI training and inference systems and helps the team develop analytical models that project end-to-end system performance. You will work across serving strategies, parallelism, collective communication, and interconnect behavior to find the architectural bottlenecks that limit performance and recommend optimal system configurations for Intel architecture.

A self-driven, hands-on, learning-by-doing mindset is essential, along with the ability to pick up unfamiliar workloads, tools, and architectures quickly.

Key Responsibilities

• Enable new AI workloads in analytical simulators, and model emerging multi-GPU deployment, parallelization, and communication strategies as simulator capabilities.
• Build analytical models of newly proposed GPU architectures, covering the GPU core, memory hierarchy, and training/inference deployment scenarios, and produce performance estimates that expose architectural bottlenecks and drive improvement recommendations.
• Integrate and modify open-source simulators or analytical models for Intel-specific architecture.
• Perform what-if analysis by sweeping core count, memory/cache capacity and bandwidth, fabric bandwidth, GPU interconnect, and deployment options, then deliver optimization and architecture recommendations.

This role is eligible for Intel's hybrid work model, which allows employees to split their time between on-site work at their assigned Intel site and off-site work. Job posting details such as work model, location, or time type are subject to change.