moffett ai

Operating Duty:

  • Be responsible for the research of frontier model compression algorithm (quantization, pruning, distillation and sparseness)
  • Be responsible for the implementation of advanced model compression algorithms
  • Participate in the design and implementation of model compression training platform
  • Assist in the application of model compression algorithm in specific fields (CV, voice)
  • Explore the next generation model acceleration algorithm



  • Master degree or above in computer, electronics, communication or mathematics, doctor is preferred
  • Be familiar with Python and have the ability to quickly implement technical prototype
  • Have a good foundation of computer architecture and understand the underlying calculation of neural network
  • Understand the common CNN network structure and image recognition / detection algorithm
  • Skillfully use the common deep learning model training framework
  • (PyTorch,TensorFlow)
  • Be familiar with the latest research and technical progress of model compression
  • Be familiar with existing model compression tools (torch.fx/tensorlite/nncf/aimet/tvm/paddleslim, etc.)
  • Excellent papers and competitions are preferre


    • 401(k)
    • Dental insurance
    • Health insurance
    • Paid time off
    • Vision insurance
    • Lifeinsurance
    • Freeonsite lunch

To apply for this job email your details to