Location: Ireland (Anywhere) Remote: Yes Willing to relocate: Yes (Within Ireland)

Technologies: Python, Java (Core), Shell Scripting, PyTorch, Tensorflow, SQL/Postgres, MongoDB

Résumé/CV: https://tejus.dev/resume.pdf

Github: https://github.com/tejus-vignesh

Portfolio: https://tejus.dev

Email: tejus.vignesh@gmail.com

Location: Ireland (Anywhere)

  Remote: Yes

  Willing to relocate: Yes (Within Ireland)

  Technologies: Python, Java (Core), Shell Scripting, PyTorch, Tensorflow, SQL/Postgres, MongoDB

  Résumé/CV: https://tejus.dev/resume.pdf

  Github: https://github.com/tejus-vignesh

  Portfolio: https://tejus.dev

  Email: tejus.vignesh@gmail.com

I am a multidisciplinary engineer with 3+ years of experience at the largest bank in India, developing payload-encrypted APIs. I am also into creating and training deep learning models from scratch, including a 60-million-parameter Transformer Model for code summarization. I have worked with LLMs like LLAMA-2 and Mistral AI by quantizing weights and using Parameter Efficient Fine-tuning methods.

Can someone explain how this will affect any open-source or transformers-based model?

It means if Google sued (not in all cases), they can block your service from using Transformer model.

Looking over the patent, I think they would really struggle to enforce this against someone using a decoder only transformer. Which is basically everyone at this point.

Yes, but patent US11556786B2 would perhaps give them that ability.

The party is over.