Skip to content
View ved1beta's full-sized avatar
🍊
santra
🍊
santra

Block or report ved1beta

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
ved1beta/README.md

image

Topics I Covered

  • Triton: making custum triton kernals for better optimizations , working on some big kernal projects
  • Cuda: cuda archirecture for better understanding of kernals and triton
  • Deep Learning: comp vision , NLP ect : )

Technical Skills πŸ› οΈ

  • Languages: Python, C , javascript
  • Frameworks & Libraries: Fast ai , Pytorch , Pandas , Matplotlib , triton
  • Tools & Platforms: GitHub, Docker, , Vercel, Neovim , Vscode , JupyterNoterbook , Aws
  • Machine Learning Specialist: Proficient in statistical analysis, predictive modeling (Regression, Decision Trees, Random Forest), and advanced algorithms (CatBoost, SGD) with strong focus on optimization and accuracy.

Key Projects πŸ“š

  • WED BOT: A bot who mimics my personality and talks like me (fun fact: It has 2 personalities). None is like the real me

  • Pdf_summariser: Innovative RAG web app for PDF analysis using machine learning. Enables intelligent document summarization and information retrieval through advanced embedding and tokenization techniques. Transforms PDF content into actionable insights with AI-powered processing.

  • LLM tokenizer: Bulding a tokenizer for llm from scratch :)

  • Loan approval predictor: Utilized a huge dataset of loan defaults and performed data transformation and model training. Tried different methods to get 95%+ accuracy and deployed with AWS and implemented CI/CD pipelines

  • Image classification Model: Trained a neural network from scratch in pure Python (no libraries) and achieved 80%+ accuracy on the MNIST dataset

  • v0 clone: made backend of v0 that for given prompt gives back code for website it writes in react and node mainly but you can choose any framework please check out ,

  • Research-copilot: working on a chat bot that will help in research and provide relavent links citations and summarization of given paper made with RAG and llama models used vector DB storage nad a=Advanced Rag models

  • Msx_gen: Bot that takes in a video extracts its mood and vibe and generated music based on the video just a curious project learned about music theory and vidoe processing from OpenCV and deploying on AWS Transformer: writing transformer from scratch : )

Working on

Many in deployment many failed T_T"

Leadership & Advocacy 🌟

  • FreecodeCamp: Wrorked as Teaching Assistance Utilized advanced teaching methods to facilitate hands-on coding projects, resulting in a 30% increase in student engagement and participation levels.

Connect with Me πŸ“¬

I’m looking forward to collaborating on projects that are at the intersection of technology and social good. Let’s connect! 🌍

Pinned Loading

  1. wed-bot wed-bot Public

    A bot with gemma api that talks likes wrote in streamlit

    Python 2

  2. v.o_clone v.o_clone Public

    wrote v0.dev backend and inspired by bolt.new in TS

    TypeScript 2

  3. LLM_chatGPT LLM_chatGPT Public

    Trained a GPT model from scratch on shakespeare's plays just generates random Play scripts it will need fit it and some post processing to deploy

    Python 2

  4. token token Public

    wrote tokenizer used in Gpt 2 from scratch

    Jupyter Notebook 2

  5. research_copilot research_copilot Public

    Research Summarizer: A RAG-based model that simplifies complex topics into clear explanations while providing accurate citations for reference.

    Python 1

  6. msx_gen msx_gen Public

    video to music

    Python 1