Skip to content
View nikitakapitan's full-sized avatar
4 followers · 3 following

Block or report nikitakapitan

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
nikitakapitan/README.md

Hi there 👋 my name is Nick, I am an AI research engineer specializing in Natural Language Processing.

Setup widget

  • 🔭 I’m currently working on LLM benchmark Ecosystem
  • 👯 I’m looking to collaborate on applied NLP to industrial projects
  • ⚡ You can also check more cool visualizations of my github projects at my website www.kapitan.uk

Here are some of my notable projects:

  • 2019 I implemented from scratch the 🔙backpropagation algorithm and 🤖Transformer architecture for 🔄ENG-GER machine translation.
  • 2020 I replicate from scratch 📐GPT-2 architecture, and 🖥️trained it on 8XA100 GPU cluster to achieve the same OpenAI score.
  • 2021 I implemented a 🌍Multi-language XML-RoBERTa for 🏷️Named Entity Recognition
  • 2021 I implemented the 🐎Pegasus model for 📝Summarization of people's 💬dialogues
  • 2021 Given a user's ❓question, I extracted relevant 📂docs and implemented miniML to answer the question
  • 2022 I implemented 🎓Knowledge distillation to reduce the 🍃model size while maintaining the same performance
  • 2023 I implemented 🎯quantization and ✂️weights pruning to reduce model 💾size and ⌛latency
  • 2023 I developed the automatic report for any LLM and any dataset to measure its 📊performance, 💽disk space and ⌛latency

Pinned Loading

  1. llama llama Public

    Forked from meta-llama/llama

    Inference code for LLaMA models

    Python 1

  2. nanoGPT nanoGPT Public

    Forked from karpathy/nanoGPT

    The simplest, fastest repository for training/finetuning medium-sized GPTs.

    Python 1

  3. nlphub nlphub Public

    NLP with Transformers using Hugging Face

    Jupyter Notebook 5 1

  4. transflate transflate Public

    baby Encoder-Decoder Transformer for GER-ENG machine translation

    Jupyter Notebook 1

  5. transformers transformers Public

    Forked from huggingface/transformers

    🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

    Python

  6. modelcraft modelcraft Public

    Fine tune or distill your model in one click

    Python