Skip to content
View calvin-fei's full-sized avatar
  • Cupertino, CA
  • 02:46 (UTC -07:00)

Highlights

  • Pro

Block or report calvin-fei

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
calvin-fei/README.md

πŸ‘‹ Hello, welcome to my repo!

Calvin Fei is a Machine Learning Engineer at AWS Annarpurna Labs, with expertise in Large Language Models (LLM), Deep Neural Networks (DNN), and High Performance Computing (HPC).

Calvin Fei is a Machine Learning Engineer specializing in large-scale ML systems, LLM inference infrastructure, and accelerator-aware model optimization. He holds a Master’s degree in Computer Science from Cornell University.

At AWS Annapurna Labs, Calvin works on AI/ML systems prototyping and performance optimization, collaborating with strategic customers to port and optimize large-scale deep learning models on AWS Trainium and Inferentia. His work focuses on LLM inference systems, model efficiency, and hardware-aware optimization, enabling teams to deploy high-performance models on specialized AI accelerators.

His expertise spans the full LLM systems stack β€” from model architectures and inference pipelines to accelerator-level kernel optimization. He has extensive experience optimizing throughput, latency, and hardware utilization for production-scale ML workloads using PyTorch, CUDA, Triton, and C/C++, and has worked with modern high-performance serving frameworks such as vLLM and TensorRT-LLM.

Calvin is particularly interested in advancing scalable LLM systems, high-performance inference infrastructure, and next-generation AI compute platforms, bridging model architectures, ML systems software, and accelerator hardware.

πŸ‘¨β€πŸ’» Open to Roles

  • Machine Learning Engineer
  • Applied Scientist

πŸ€– Projects - Large Language Models (LLM)

πŸ“ˆ GitHub Stats

Cheng's Current Streak

Profile Views stars

Pinned Loading

  1. mini-torch mini-torch Public

    Mini version of PyTorch, built from scratch with Python, Numba, CUDA.

    Python

  2. text-transformer text-transformer Public

    Python

  3. llm-driven-red-teaming llm-driven-red-teaming Public

    Automated red teaming generator driven by LLM.

    Python

  4. cnn-driven-architecture-style-classifier cnn-driven-architecture-style-classifier Public

    Architecture style classifier driven by CNN, built with PyTorch and ResNet.

    Jupyter Notebook

  5. dqn-driven-route-planner dqn-driven-route-planner Public

    Google Maps route planner driven by Deep Q-Networks, built with Python and Tensorflow.

    Python