Neural Network Compression Engineer

Job Title

Neural Network Compression Engineer

Role Summary

Develop and maintain neural network compression algorithms and tooling for Ambarella's AI vision edge processors. The role focuses on optimizing models for resource-constrained devices used in camera and vision applications.

Experience Level

Mid-level. No specific years-of-experience requirement stated.

Responsibilities

Primary responsibilities include:

• Develop and maintain neural network compression algorithms and tools for deployment on edge processors.
• Design and apply pruning, quantization, and knowledge distillation techniques to reduce model size and latency.
• Evaluate compressed models for accuracy, latency, and resource usage on target hardware.
• Collaborate with software and hardware teams to integrate optimized models into inference pipelines.

Requirements

Key technical requirements:

• Must-have: Strong knowledge of neural network compression methods, including pruning, quantization, and knowledge distillation.
• Must-have: Strong background in CNN and transformer-based computer vision and large language model concepts.
• Must-have: Familiarity with computer vision and machine learning frameworks (e.g., PyTorch, TensorFlow).
• Must-have: Proficient in C/C++ and/or Python programming.
• Nice-to-have: Experience with compression-aware training (pruning/quantization-aware training) and model optimization for edge deployment.

Education Requirements

Not specified.

About the Company

Company: Ambarella

Headquarters: Santa Clara, California, USA

Ambarella is a leader in computer vision and video processing technology, providing advanced solutions that enhance the performance of video applications. Focused on quality and innovation, Ambarella develops products for diverse uses, including autonomous vehicles, surveillance systems, and smart cameras, aiming to deliver pristine imagery and efficient compression while minimizing power consumption.