Intel Off Campus Drive 2026 | AI System Performance Engineer

Intel is hiring an AI System Performance Engineer for its Bangalore team, with the role suited to college graduates and candidates with experience in AI workload and GPU architecture modelling or simulation. Based at SRR4 – Sarjapur 4, the engineer will analyse AI training and inference systems, model multi-GPU deployments and identify architectural bottlenecks across Intel platforms. The work covers GPU architecture, memory hierarchy, interconnects, serving strategies and large language model workloads. It is a strong role for candidates who want hands-on exposure to analytical simulators, emerging AI systems and performance optimisation at architecture level.

Company Overview

Intel is a global semiconductor company that develops processors, graphics technologies, data-centre products and platforms for computing and artificial intelligence. Its engineering teams work across CPU and GPU architecture, software, manufacturing and performance optimisation, making Intel experience relevant for candidates building careers in computer architecture and AI systems.

Role Overview

The role focuses on predicting and improving end-to-end performance for AI training and inference systems. The engineer will use analytical models and simulators to assess Intel architecture, compare deployment choices and recommend configurations that address performance limits.

• Build an understanding of AI training and inference systems and their performance behaviour.
• Model serving strategies, parallelism, collective communication and interconnect behaviour.
• Study performance across multi-GPU deployments and different AI workload configurations.
• Use analytical results to identify architectural bottlenecks and support system-level recommendations.

Eligibility & Qualifications

• BTech or Master’s degree in Electrical Engineering, Computer Engineering, Computer Science or a related field.
• College graduates can apply.
• Candidates with 1+ years of experience in AI workload and GPU architecture modelling or simulation, or a similar field, can apply.
• Strong proficiency in Python for developing modular, maintainable analytical models and simulation frameworks.
• Understanding of computer architecture, including GPU and CPU components, ALU/SM cores, caches, memory types and memory hierarchy.
• Understanding of LLM AI workloads such as Transformer, MoE, Attention, MLA, LLaMA, DeepSeek, Qwen and robotic workloads, along with kernel internals.

Key Responsibilities

• Enable new AI workloads in analytical simulators.
• Model emerging multi-GPU deployment, parallelisation and communication strategies as simulator capabilities.
• Build analytical models for proposed GPU architectures, covering the GPU core, memory hierarchy and training or inference deployment scenarios.
• Produce performance estimates that expose architectural bottlenecks and support improvement recommendations.
• Integrate and modify open-source simulators or analytical models for Intel-specific architecture.
• Perform what-if analysis by varying core count, memory and cache capacity, bandwidth, fabric bandwidth, GPU interconnect and deployment options.
• Deliver optimisation findings and recommendations for suitable system architectures.

Skills Required

• Python programming and analytical model development.
• Computer architecture knowledge covering GPUs, CPUs, ALU/SM cores, caches and memory hierarchy.
• Knowledge of LLM workloads, kernel internals and AI training and inference mechanics.
• Understanding of distributed training and inference, including data, pipeline, tensor and expert parallelism.
• Knowledge of communication patterns when scaling workloads across multi-GPU hardware.
• Understanding of prefill and decode, KV cache, batching policies and quantisation.
• Exposure to TRT-LLM, vLLM or SGLang.
• Familiarity with Git/GitHub, CI/CD and automated workflows.
• Understanding of power fundamentals and power-performance co-optimisation.
• A self-driven, hands-on learning-by-doing mindset.
• Ability to learn unfamiliar workloads, tools and architectures quickly.

How to Apply for Intel Off Campus Drive 2026

Eligible and Interested candidates can apply using the below-mentioned link before the link expires. Applicants need to register with the company portal in order to complete the registration process.

Apply Link: Click Here To Apply