Expoint - all jobs in one place

The point where experts and best companies meet

Limitless High-tech career opportunities - Expoint

Qualcomm Machine Learning Performance Engineer - Video Codec Optimization 
Germany, Bavaria, Munich 
935015217

14.08.2024

Job Area:

Engineering Group, Engineering Group > Machine Learning Engineering

Qualcomm's Multimedia R&D Standards Group is seeking a skilled C++ Machine Learning Performance Engineer to optimize and integrate AI models into our next-generation video codecs. As video data continues to grow exponentially, we need to leverage AI to improve compression efficiency. The ideal candidate will be responsible for the end-to-end process of quantizing PyTorch models, porting them to highly optimized C++ code using SIMD and AVX2 instructions, and deploying them within traditional video codec frameworks. Experience with video processing is extremely valuable for this role. You will work with a world-class team of data compression experts to push the boundaries of AI-enabled video compression.

Responsibilities:

• Quantize and optimize PyTorch models for efficient deployment on edge devices, particularly Qualcomm's NPU.

• Port PyTorch models to highly optimized C++ implementations using SIMD and AVX2 instructions.

• Integrate optimized ML models into existing video codec frameworks.

• Collaborate with research scientists to implement and optimize novel AI-based video compression algorithms.

• Benchmark and profile ML model performance, identifying and resolving bottlenecks.

• Contribute to the development of tools and workflows for efficient model optimization and deployment.

• Document optimization techniques, best practices, and integration processes.

• Strong C++ programming skills with a focus on performance optimization.

• Extensive experience with SIMD instructions (SSE, AVX, AVX2, NEON) and vectorization techniques.

• Proficiency in low-level optimization techniques and parallel programming.

• Experience with PyTorch and model quantization techniques.

• Solid understanding of video processing and compression techniques.

• Experience with one or more modern video codecs (VVC or AV1 experience is a huge plus).

• Knowledge of computer architecture and its impact on performance.

• Familiarity with profiling tools and performance analysis.

• Experience optimizing ML models for edge deployment, particularly on mobile or embedded platforms.

• Strong problem-solving skills and attention to detail.

• Excellent written and verbal communication skills.

Minimum Qualifications:

• BS or MS in Computer Science, Electrical Engineering, or a related field.

• 3+ years of experience in C++ development and performance optimization.

• Demonstrated experience in porting and optimizing ML models for deployment.

• Familiarity with video codecs and compression techniques.

Preferred Qualifications:

• Experience with VVC or AV1 codec implementation or optimization.

• Track record of contributions to open-source video processing or codec projects.

• Publications or patents related to video compression or ML optimization.

Minimum Qualifications:

• Bachelor's degree in Computer Science, Engineering, Information Systems, or related field and 4+ years of Hardware Engineering, Software Engineering, Systems Engineering, or related work experience.

Master's degree in Computer Science, Engineering, Information Systems, or related field and 3+ years of Hardware Engineering, Software Engineering, Systems Engineering, or related work experience.

PhD in Computer Science, Engineering, Information Systems, or related field and 2+ years of Hardware Engineering, Software Engineering, Systems Engineering, or related work experience.

Qualcomm expects its employees to abide by all applicable policies and procedures, including but not limited to security and other requirements regarding protection of Company confidential information and other confidential and/or proprietary information, to the extent those requirements are permissible under applicable law.

Pay range:

$148,500.00 - $222,500.00