AI Developer

Job Overview

The role involves advancing state-of-the-art in Machine Learning using Qualcomm chips' heterogeneous computing capabilities. This allows the running of trained neural networks on device without a cloud connection. The inference engine enables developers to run neural network models trained in various frameworks on Snapdragon platforms at high speeds while consuming minimal power. Your work will directly impact billions of mobile devices worldwide.

In this position, you will be responsible for developing and commercializing ML solutions like Snapdragon Neural Processing Engine (SNPE)/Android Neural Network framework on Qualcomm SoCs. You will port ML solutions to various platforms involving different compiler toolchains and enable the product to work on upcoming chipsets. Performance optimization for benchmarking on multiple hardware accelerators (like CPU/GPU/DSP) is also a key responsibility. You will possess expert knowledge of debugging, improving, and maintaining cross-platform compilation, linkage, build system, and deployment aspects of large software C/C++ dependency stacks using best practices.

You will also need to stay up-to-date with the fast-paced development happening in the industry and academia to continuously enhance our solution from software engineering as well as machine learning perspectives.

Key Responsibilities:

• Develop and commercialize ML solutions like SNPE/Android NN framework on Qualcomm SoCs
• Port ML solutions to various platforms involving different compiler toolchains
• Enable the product to work on upcoming chipsets
• Optimize performance for benchmarking on multiple hardware accelerators (CPU/GPU/DSP)
• Stay updated with industry and academia developments

Required Skills and Qualifications:

To excel in this role, you should have a strong background in quality software development with excellent analytical and debugging skills. Very good development skills in C, C++, STL, and Python are essential. Experience with embedded systems development or equivalent is required. A good understanding of software development and debugging in embedded environments is necessary. Additionally, experience in analysis and optimizations of ML KPIs like Performance, Accuracy metric, and system KPIs like Processor utilization, Memory, Cache, DDR etc., is highly desirable. Familiarity with version control tools like git, Gerrit, is also required. Excellent communication skills (verbal, presentation, written) and the ability to represent complex data in a well-organized manner are crucial. Collaborating across a globally diverse team and multiple interests is also an essential aspect of this role.

Preferred Qualifications:

A good understanding of Machine Learning and Deep Learning fundamentals and key concepts is beneficial. Experience with ML application development or any benchmarking framework like MLPerf, AITUTU, GBAI etc. is desirable. Familiarity with ML frameworks like TensorFlow, PyTorch, Onnx etc. can be advantageous. Awareness of windows ecosystem and familiarity with windows debugging/analysis tools can also be helpful. Good understanding of Processor architecture, system design fundamentals, and experience with machine learning accelerators or GPU/DSP development, optimizing algorithms for hardware acceleration cores using OpenCL, DSPs and/or ML specific cores are preferred qualifications.

Work Experience:

2 to 4 years of relevant work experience in software development and optimizations are expected.

Educational Requirements:

Bachelor's/Master's in Computer Science, Computer Engineering, or Electrical Engineering are the minimum educational requirements.