AI Inference Performance Engineer

QUALCOMM, Inc.; Cork; On-site; EUR 90,000 - 150,000; Full time

Benefits offered by this job

Stock options

Performance bonus

Relocation support

Healthcare

Job summary

Qualcomm Ireland is seeking an AI Performance Engineer to join its Cloud AI Engineering group in Cork. You will work across the full product lifecycle from R&D to deployment, converting and optimizing models for efficient inference using PyTorch and ONNX.

The role emphasizes transformer architectures, performance analysis, and collaboration with internal teams to scale workloads on current and future hardware.

Qualifications

• Bachelor's or higher in Engineering/CS or related field with several years in software/ML.
• Proven experience building/optimizing language models in production environments.
• Strong knowledge of transformer architectures and accuracy-performance trade-offs.

Responsibilities

• Convert, optimize and deploy models for efficient inference using PyTorch and ONNX.
• Analyze and optimize LLM/VLM/diffusion models for throughput and latency.
• Map AI workloads to current and future hardware designs with parallelism strategies.
• Collaborate with compiler, firmware and platform teams to deliver solutions.
• Develop high-level kernels, e.g. in Triton, for efficient low-level code.

Skills

Python

PyTorch

ONNX

Transformers

Performance optimization

C++

Java

Deep Learning

Education

MS in CS/ML

PhD in CS/ML

Tools

Triton

torch.compile

Job description

Qualcomm Ireland is seeking an AI Performance Engineer to join its Cloud AI Engineering group in Cork. You will work across the full product lifecycle from R&D to deployment, converting and optimizing models for efficient inference using PyTorch and ONNX.

The role emphasizes transformer architectures, performance analysis, and collaboration with internal teams to scale workloads on current and future hardware.