Skip to content
View zyberg2091's full-sized avatar
🎯
Focusing
🎯
Focusing
16 followers · 50 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Highlights

Block or report zyberg2091

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
zyberg2091/readme.md

Hey there

I am Shubham Kumar.

AI/ML Developer • GenAI Enthusiast • Research-Oriented Engineer

I build and explore machine learning systems with a focus on multilingual NLP, LLMs, and real-world AI applications.

🔗 Connect with me

🌍 Open Source Contributions

  • 🧠 Hugging Face Transformers
    🔗 huggingface/transformers#9286

    • Contributed to resolving a token alignment issue affecting multilingual NER pipelines
    • Improved tokenizer behavior across languages

  • 🗣️ Hugging Face Model Hub
    🔗 https://huggingface.co/zyberg2091/distilbert-base-multilingual-toxicity-classifier

    • Published a multilingual toxicity classifier (DistilBERT)
    • Supports English, Hindi, and Hinglish
    • Focused on real-world content moderation use cases

  • 📊 TensorFlow Models (Community Contributions)

    • Contributed to discussions on best practices for object detection on custom datasets

  • 📚 Documentation Contributions (TensorFlow, Rasa)

    • Improved developer experience through documentation fixes and clarity enhancements

📊 GitHub Stats

Pinned Loading

  1. LocalLLM-Design LocalLLM-Design Public

    A Small Pretrained Language Model built with research and learning purpose

    Jupyter Notebook

  2. gated-attention-dynamics gated-attention-dynamics Public

    Study Gated Attention for Language Models Under Different Normalization Techniques: Nonlinearity, Sparsity, and Attention Sink Suppression

    Jupyter Notebook

  3. e-safety e-safety Public

    An architecture designed to predict toxicity of web pages, adult video embedded in web pages.It automatically blocks the adult video if any content in the video represents child pornography.

    Jupyter Notebook

  4. BlogTagger BlogTagger Public

    This is a package that can extract important keywords from a page to create tags for any blogs, news etc. The link to the source repository of package is mentioned in the readme.

    Python

  5. Aapka-Apna-Hiphop Aapka-Apna-Hiphop Public

    Make your own rap verses with one click

    Jupyter Notebook 10 3