Engineer – Voice AI Systems Machine...

About Qualcomm India Private Limited

Qualcomm India Private Limited is seeking a talented and highly motivated Engineer specializing in the implementation of Voice AI and Audio solutions. This role is within the Engineering Group, specifically in Systems Engineering.

Job Summary

The Audio Systems team at Qualcomm is looking for an engineer to prototype, optimize, and productize state-of-the-art Machine Learning (ML) models for Voice AI and Audio applications. The focus will be on ensuring efficient deployment of these models on Snapdragon platforms.

Key Responsibilities

• Develop, optimize, and deploy Voice AI and audio ML models, prioritizing inference efficiency across Neural Processing Units (NPUs), Graphics Processing Units (GPUs), and Central Processing Units (CPUs).
• Conduct model evaluation, quantization, and compression to achieve fast and robust inference on embedded hardware.
• Analyze and compare various ML model architectures, including Diffusion Models, U-Nets, Transformers, BERT, and BART, for suitability in audio applications.
• Collaborate with cross-functional R&D, systems, and integration teams to verify system use cases and provide commercialization support.
• Contribute to the design and software implementation of audio ML models using embedded C/C++ and Python.
• Evaluate and debug system performance, optimizing for both speed and robustness.
• Stay abreast of industry trends, perform benchmarking and performance analysis of different model architectures, and introduce innovative architectural or technical advancements to the team.

Required Skills and Experience

• Proficiency in C/C++ and Python programming.
• Demonstrated experience with audio processing and embedded solutions.
• Hands-on experience working with audio frameworks and audio solutions on various platforms.
• Familiarity with ML frameworks such as PyTorch, TensorFlow, and ONNX.
• Knowledge of model quantization and compression techniques, with practical experience in optimizing inference and deployment on embedded hardware.
• Strong understanding of ML model architectures including CNNs, RNNs, Transformers, and U-Nets, as well as statistical modeling techniques.
• Understanding of Digital Signal Processor (DSP) or Microcontroller architectures and frameworks.
• Experience in developing and debugging software on embedded platforms, including familiarity with software design patterns, multi-threaded programming (e.g., POSIX, PTHREADS), and fixed-point coding.
• Excellent verbal and written communication skills, with the ability to work effectively both independently and as part of geographically dispersed, multidisciplinary teams.
• Proven ability to thrive in a dynamic, multi-tasking environment, demonstrating quick learning, self-motivation, and a results-driven approach.

Minimum Qualifications

• Bachelor's, Master's, or PhD in Computer Science, Electronics and Communication, Electrical Engineering, or a related field (or equivalent practical work experience).
• Bachelor's degree in Engineering, Information Systems, Computer Science, or related field and 2+ years of Systems Engineering or related work experience.
• OR Master's degree in Engineering, Information Systems, Computer Science, or related field and 1+ year of Systems Engineering or related work experience.
• OR PhD in Engineering, Information Systems, Computer Science, or related field.

Preferred Qualifications

• Experience with Qualcomm AI hardware accelerators (NPUs) and Qualcomm Software Development Kits (SDKs).
• Knowledge of Qualcomm's Machine Learning framework (QNN), platforms, and tools.

Keywords

Audio ML Models, RNNs, Audio dataset, Inference Optimization, ML model Quantization, Quantization aware, post training quantization, Model optimization