Intel has announced an opening for an AI System Performance Engineer (Req JR0288211) based in Bangalore. This technical position focuses on benchmarking, characterizing, and optimizing artificial intelligence workloads on next-generation semiconductor architectures.
Role Overview
In this role, you will work at the critical intersection of silicon hardware and modern machine learning software stacks. Intel's AI systems teams ensure that large-scale neural networks, including vision transformers and generative LLMs, achieve maximum throughput and energy efficiency across Intel Xeon CPUs, Gaudi accelerators, and discrete GPUs.
What the role involves
Your responsibilities will center on profiling AI model execution to identify memory bandwidth limits, compute bottlenecks, and latency stalls. You will run performance simulations, write automated benchmarking suites in Python and C++, and analyze hardware performance counters using tools like Intel VTune. You will collaborate closely with compiler engineers and kernel developers to optimize matrix multiplication kernels, attention layers, and distributed communication primitives across multi-node clusters.
Who it suits
This role suits engineering graduates with a degree in Computer Science, Electronics, Electrical Engineering, or related disciplines who have strong academic training in computer architecture, cache hierarchies, and parallel computing. If you love understanding how software instructions execute on silicon and enjoy squeezing maximum performance from hardware accelerators, this role offers unmatched learning opportunities.
How to stand out
Highlight coursework or personal projects that involve CUDA, OpenCL, SIMD intrinsics, or hardware modeling. Experience using performance profilers (VTune, Nsight, or perf) and an understanding of roofline performance models will distinguish your candidature immediately. Be ready to discuss trade-offs between precision formats (FP32, BF16, FP8) and memory bandwidth saturation during technical interviews.
Check before you apply
Intel does not disclose salary ranges in the job notification, but performance engineering positions at Intel Bangalore typically command compensation between 12 to 16 LPA for early-career hires. The position is situated in Bangalore with on-site lab access required for hardware testing.
FAQs
Who can apply for this position?
Recent graduates and postgraduates (B.Tech, M.Tech, MS) in CS, ECE, or EE with interest in hardware-software co-design can apply.
Which programming languages are essential?
Proficiency in C, C++, and Python is vital, alongside familiarity with Linux systems and shell scripting.
Where is the job based?
The role is located at Intel's campus in Bangalore, Karnataka.
