Job Description
The engineer will be responsible for the design and development of optimization tools for neural networks, transformers, and large language models (LLMs). This role involves applying post-training, training-aware, and other advanced optimization techniques to enhance model efficiency and performance.
Key responsibilities:
1. Develop optimization toolchains for computer vision models and ISP models (denoising, super-resolution, detail enhancement, etc.).
2. Perform hardware-aware model optimization and porting for Ambarella platforms.
3. Research and evaluate emerging technologies, including pruning, quantization, and fine-tuning techniques for convolutional neural networks (CNNs), transformers, and LLMs.
4. Provide technical support and solutions to customers regarding model optimization and deployment.
Job Requirements:
1. Experience of model deployment would be a plus.
2. Experience of model optimization such as pruning and quantization would be a plus
3. Experience of ISP model fine-tuning would be a plus
4. Experience of model porting would be a plus
5. Background of machine learning based computer vision or ISP knowledge
6. Familiar with machine leaning frameworks like Pytorch, TensorFlow, or Huggingface
7. Skilled in Python and C/C++ programing.
Skills Required
- Background in machine learning based computer vision or ISP knowledge
- Familiar with machine learning frameworks like PyTorch, TensorFlow, or Hugging Face
- Skilled in Python and C/C++ programming
- Experience of model deployment
- Experience of model optimization such as pruning and quantization
- Experience of ISP model fine-tuning
- Experience of model porting
What We Do
Ambarella (Nasdaq: AMBA) is a leading developer of visual AI products. Our technologies enable a wide variety of human and computer vision applications, including video security, advanced driver assistance systems (ADAS), electronic mirror, drive recorder, driver/cabin monitoring, autonomous driving, and robotic applications. Ambarella’s low-power system on chips (SoCs) offer high-resolution video compression, advanced image processing, and powerful deep neural network processing to enable intelligent cameras to extract valuable data from high-resolution video streams. For more information, please visit www.ambarella.com
.jpeg)
