Senior AI Quantization & On-Device Optimization Engineer

QUALCOMM, Inc.; San Diego (CA); Hybrid; USD 141,000 - 211,000; Full time

Job summary

Qualcomm Technologies, Inc. seeks a Senior AI Research Quantization Engineer in San Diego to advance on-device generative AI, LLMs, and multi-modal models, collaborating across hardware, software, and systems to deliver state-of-the-art efficiency.

You will prototype AI systems, develop efficient inference methods, and explore quantization, model compression, and neural search, applying Python and PyTorch to push performance within power and memory constraints across devices and cloud platforms.

Qualifications

• Bachelor's, Master's, or PhD in Computer Science, Engineering, Information Systems, or related field.
• 2+ years of hardware, software, or systems engineering experience or 1+ year with Master's/PhD.
• Experience with machine learning algorithms or systems engineering.

Responsibilities

• Research and develop algorithms for efficient generative AI, LLMs, LVM, and multimodal models.
• Design and implement efficient inference techniques (batching, KV caching, attention optimizations, long context, speculative decoding).
• Develop advanced quantization algorithms for complex models and hardware-in-the-loop evaluation.
• Explore model compression and neural search; prototype Generative AI systems; enable device and cloud deployments.

Skills

Python

PyTorch

Education

Bachelor's degree in CS or related

Master's degree in CS or related

PhD in CS or related

Tools

Python

PyTorch

Job description

Qualcomm Technologies, Inc. seeks a Senior AI Research Quantization Engineer in San Diego to advance on-device generative AI, LLMs, and multi-modal models, collaborating across hardware, software, and systems to deliver state-of-the-art efficiency.

You will prototype AI systems, develop efficient inference methods, and explore quantization, model compression, and neural search, applying Python and PyTorch to push performance within power and memory constraints across devices and cloud platforms.