Shorya Sethia
shoryasethia
Final year student at IIT Bombay | Expect the Unexpected
Languages
Repos
44
Stars
232
Forks
22
Top Language
Jupyter Notebook
Loading contributions...
Top Repositories
A Python package for converting PDFs to markdown while extracting images and tables, generate descriptive text descriptions for extracted tables/images using several LLM clients. And many more functionalities. Markdrop is available on PyPI.
SoC [ IIT Bombay ] : Gesture based text creation [ Project ID: 102 ]
AI-Powered Financial Document Analysis Tool
Backend for YouTube inspired platform
Includes vanilla, dc, de-noising autoencoders
Project explores the evolution of image generation, from autoencoders, GANs to SD & the new VAR (Visual Autoregressive Representation) method. VAR enhances image quality and scalability beyond diffusion models. We’ll analyze each stage, explore unique emerging techniques, and work toward developing a novel approach in the near future.
Repositories
44A Python package for converting PDFs to markdown while extracting images and tables, generate descriptive text descriptions for extracted tables/images using several LLM clients. And many more functionalities. Markdrop is available on PyPI.
SoC [ IIT Bombay ] : Gesture based text creation [ Project ID: 102 ]
An agentic approach for querying and analyzing UPI transaction data using natural language
Backend for YouTube inspired platform
Regulatory intelligence that maps how rules evolve. Structured reference extraction for SEBI compliance
[AI-ML-GC] Automated investment teaser generation from company data — editable PowerPoint decks with native charts, anonymization, and full source citations.
Blogify is full-featured blogging platform built with React, Appwrite, TinyMCE, and React Hook Form. It allows users to create, edit, and publish textual blog posts with images easily
A modular voice assistant application for experimenting with state-of-the-art transcription, response generation, and text-to-speech models. Supports OpenAI, Groq, Elevanlabs, CartesiaAI, and Deepgram APIs, plus local models via Ollama. Ideal for research and development in voice technology.
Includes vanilla, dc, de-noising autoencoders
No description provided.
A machine learning project implementing multiple approaches to detect fraudulent credit card transactions.
AI-Powered Financial Document Analysis Tool
Passive Captcha Defense based on behavioural analytics and honeypot Traps with advanced ML Models for Bot Detection
A comprehensive sentiment analysis system that combines text-based machine learning with emoji sentiment scoring to classify tweets as positive or negative.
Agentic IDE build using monaco, pyodide and langGraph.
Project explores the evolution of image generation, from autoencoders, GANs to SD & the new VAR (Visual Autoregressive Representation) method. VAR enhances image quality and scalability beyond diffusion models. We’ll analyze each stage, explore unique emerging techniques, and work toward developing a novel approach in the near future.
React tutorials and projects
In Progress
An intelligent chatbot that helps you understand and navigate your codebase using LiteLLM. Ask questions about your code in natural language and get detailed, context-aware responses.
Javasript tutorials, basic problems and few projects
Contains my langGraph learnings
Bidirectional chat app and simple file transfer using Socket Programming in C++ using the Winsock2 API on Windows.
Qualcomm VisionX, Team Name : ClarifyAI, Rank : 4
Workflows and project built for hizen.ai
Comparison between several Simple Diffusion models. Used their inference client with safety_check = None
CE6001 Course Project
No description provided.
Streamlit Interface for determining the time complexity of any code based on Gemini's Response
Implemented a paper explaining Block Switching in Deep Learning on CIFAR10
Tried several ML and DL models for 0-9 Digit Classification using MNIST Dataset