Senior Software Engineer, AI Benchmarks Software

As a leading technology innovator, Qualcomm pushes the boundaries of what's possible to enable next-generation experiences and drive digital transformation, creating a smarter, connected future for all. As a Qualcomm AI Software Engineer, you will develop and implement cutting-edge machine learning techniques that enable the efficient utilization of state-of-the-art solutions across various technology verticals.

In this position, you will be responsible for integrating the Qualcomm AI SDK into AI benchmark applications across multiple hardware accelerators, including NPU, CPU and GPU, and achieving competitive benchmark scores through targeted performance and accuracy optimizations.

You will have the opportunity to demonstrate your passion for software design and development through your analytical, design, programming, and debugging skills.

Responsibilities

• Integrate the Qualcomm AI SDK into AI benchmarking applications and deploy ML models — including computer vision, language, and generative AI — on Snapdragon hardware accelerators, including NPU, CPU and GPU.

• Design, evaluate, and optimize AI benchmark model’s performance and accuracy

• Collaborate closely with AI SDK, NPU hardware accelerator, CPU scheduling, and power/performance teams to conduct end-to-end performance analysis and implement optimizations that deliver industry-leading benchmark scores.

• Engage with AI benchmarking communities and vendors to discuss and propose enhancements to scoring methodologies and accuracy metrics, as well as to advocate for the inclusion of new, representative models.

• Analyze competitor’s SDKs used in AI benchmark applications to develop strategies that strengthen Qualcomm’s competitive position and benchmark leadership

• Contribute to a culture of technical excellence, knowledge sharing, and continuous improvement within the AI Software team.

• Work independently with minimal supervision; Staff Engineers provide guidance to junior team members and make decisions impacting their area and the broader work group

Preferred Qualifications:

• Master’s degree in Computer Science, Computer Engineering, or Electrical Engineering

• 1+ years (Senior) / 3+ years (Staff) of experience building embedded software applications.

• Experience with ML application development or any benchmarking app experience like MLPerf, Geekbench AI, Procyon AI, AITutu etc.

• Understanding of Machine learnings and Deep Learning fundamentals and key concepts

• Familiarity with various generative AI model architectures such as LLMs.

• Experience with development on various platforms such as Linux, Android, or Windows.

• Experience with any machine learning framework like TensorFlow, Pytorch, Onnx etc.

• Good understanding about Processor architecture, system design fundamentals. Experience with machine learning accelerators or GPU/DSP development, optimizing algorithms for hardware acceleration cores using OpenCL, DSPs and/or ML specific cores

• Experience in analysis and optimizations of ML KPIs like Performance, Accuracy metric, as well as system KPIs like Processor utilization, Memory, Cache, DDR etc.,

• Experience with SW optimizations for custom SOCs/Accelerators, SIMD, vectorization

• Familiarity with build frameworks like CMake, Make, Gradle Experience using/integrating Qualcomm AI Stack products (e.g. QNN, Genie).

• Well versed in version control tools like git, Gerrit.

• Ability to collaborate across a globally diverse team and manage multiple interests.

Qualcomm is an equal opportunity employer. If you are an individual with a disability and need an accommodation during the application/hiring process, rest assured that Qualcomm is committed