Pacman playing agent - project for RL course

A production grade ecommerce website with a recommender system pipeline

A mental wellness app for early depression detection

An AI based automated presentation coach for UDBHAV - Inter IIIT hackathon

Dual-stream deep learning architecture for deepfake detection combining spatial (RGB) and frequency-domain (DCT) analysis with CBAM attention mechanisms.

A real time chat application using MERN stack with socket.io

A starter template for building powerful, local, tool-calling LLM agents using LangGraph and llama-cpp-python

FFSD project PETVERSE, a website for everything related to pets

Benchmarking multiple recommendation models on the Instacart 2017 dataset. Starting with Factorization Machines (FM), this project will extend to deep learning approaches such as DIN, SIM, CAN, Tra…

An anomaly detection model for HDFS_v1 log dataset. Kafka to simulate real time data streaming and model retraining on new unseen data.

A basic ML model that uses linear regression to predict home prices, and the web app interface is made with Streamlit

A Chatbot for an ERP system made using rasa framework

A streamlit based application for indian farmers that gives crop recommendation, weather forecast, crop disease detection, soil analysis and steps to be taken for the crop grown and predicted yield

A ML model integrated with flask for API and html, css, javascript for the front end deployed n render with docker

Config files for my GitHub profile.

A custom Neural Network library made purely using numpy.

PPO agent and A2C agents for Flappybird. Includes scripts, training code, and evaluation tools.

Pytorch implementation of Vision Transformer from scratch on MNIST Dataset

In this i have explored different parts of an LLM from the tokenizer to the attention mechanism to predicting tokens. I have also implemented the GPT2-small (124M) architecture using pytorch

Siamese U-Net with ResNet34 encoder for pre/post-event damage classification and segmentation on xView2.

A personal RAG based chatbot that uses llama-3.1-8B-instrcut-Q4_K_M.gguf model as its llm and ChromaDB to store and repond to the user on their day to day queries. Intelligent session management wi…

A genral RAG Search chatbot, with SoTA RAG techniques such as HyDE, Hybrid retrieval with BM25 + RRF and Cross encoder reranking. Evaluated on the BEIR scifact dataset and compared all the differen…

A website hosted on github to showcase my portfolio

Jarvis like Personal Assistant for running locally in my personal laptop