Skip to content
View debashishc's full-sized avatar
Coffee please, no milk.
Coffee please, no milk.
  • AU

Highlights

  • Pro

Block or report debashishc

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, 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
debashishc/README.md

Hi there 👋

  • 🔭 I’m currently working on multi-modal machine learning and dimensionality reduction techniques.

  • 🌱 I’m learning programming Hetergenous Parallel Systems, focusing on CUDA and Triton for efficient machine learning implementations.

  • 🤔 I’m open to help with model distillation, quantization and efficient training and deployment of large ML models.

  • 👨‍💻 Research Interests

    • Developing and optimizing multi-modal model architectures with scalable data input processing to enhance performance in distributed environments.
    • Streamlining model training, fine-tuning, and inference processes for seamless integration with cloud and edge computing interfaces.
    • Implementing hardware- and IO-aware ML model compression techniques to optimize inference efficiency and reduce computational overhead.

Languages and Tools

python git aws bash bitbucket cmake cplusplus docker fastapi gcc gitlab grafana haskell latex markdown neovim numpy pytorch streamlit vim

GitHub Stats

Pinned Loading

  1. kernelheim kernelheim Public

    KernelHeim – development ground of custom Triton and CUDA kernel functions designed to optimize and accelerate machine learning workloads on NVIDIA GPUs. Inspired by the mythical stronghold of the …

    Python 2

  2. semantic-search semantic-search Public

    Implementation of semantic search using Sentence-BERT (SBERT) for a workshop. It demonstrates how to generate sentence embeddings and perform search based on cosine similarity, allowing for meaning…

    Jupyter Notebook

  3. classification-real-fake-text classification-real-fake-text Public

    Classifying real and fake text using metrics measuring human-written and machine-generated text

    Jupyter Notebook 1

  4. deep-learning-project-template deep-learning-project-template Public template

    Forked from Lightning-AI/deep-learning-project-template

    Pytorch Lightning code guideline for conferences

    Python

  5. cuda-mode-lectures cuda-mode-lectures Public

    Forked from gpu-mode/lectures

    Material for cuda-mode lectures

    Jupyter Notebook 1