Skip to content
View ayushbits's full-sized avatar
33 followers · 11 following

Achievements

Achievement: StarstruckAchievement: YOLOAchievement: Pull Shark

Achievements

Achievement: StarstruckAchievement: YOLOAchievement: Pull Shark

Block or report ayushbits

Report abuse

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

Report abuse
ayushbits/README.md

Hi there 👋

  • Currently working at NVIDIA on AI problems !

  • Previously, completed PhD from IIT Bombay (India) advised by Prof. Ganesh Ramakrishnan

  • 🔭 During PhD, I worked on semi-supervised ML problems for classification and generative usecases. Google Scholar

  • Other activities : I also developed and manage Temples of India, World's largest Hindu Temples portal constains a listing of around 7 lakh temples from across the country and detailed information about more than 5000 temples. Collaborator - Arun Jayaramakrishnan

  • Key member of UDAAN Project

  • Visit my website to know more.

Pinned Loading

  1. ParamBench ParamBench Public

    Benchmark to evaluate graduate level understanding of LLMs for Indian context in Hindi.

    Python 2 2

  2. decile-team/spear decile-team/spear Public

    SPEAR: Programmatically label and build training data quickly.

    Jupyter Notebook 110 22

  3. pe-ocr-sanskrit pe-ocr-sanskrit Public

    Source and Data of our EMNLP Paper 'A Benchmark and Dataset for Post-OCR text correction in Sanskrit'

    Python 11 3

  4. Saamayik Saamayik Public

    Source code and dataset for the paper 'Saamayik: A Benchmark and Dataset for English-Sanskrit Translation'

    Python 15

  5. llm-development llm-development Public

    This repository contains scripts and guides presented during multiple LLM development sessions by me.

    Jupyter Notebook 21 9

  6. UDAAN-LEAP/leap-pe-tool UDAAN-LEAP/leap-pe-tool Public

    A framework for assisting human while correcting the translation/OCR errors in documents, mostly dedicated to Indian Languages.

    C++ 33 10