nlp

All my experiments with the various transformers and various transformer frameworks available

Flask API for generating text embeddings using OpenAI or sentence_transformers

Kairos: An autonomous development supervisor powered by Context Engineering. It provides a living memory and a contextual constitution for your projects to prevent AI context loss.

An application that summarizes and classifies long text using zero-shot transfer-learning hosted on Huggingface spaces.

I implement a deep learning network to classify COVID-19 Tweets into 5 categories and 3 categories using DistilBERT (a lighter version of BERT) as an embedding layer along with an LSTM and Dense Layer. I Achieve 65% accuracy with 5 categories and 80% accuracy on 3 categories.

LegalEase AI is a document simplification tool built using Gemini API, Streamlit, and Hugging Face models. It allows users to upload legal PDFs and automatically receive simplified summaries, clause-level insights, and structured information designed for clarity and accessibility.

MathBot is a transformer-based Math Word Problem (MWP) solver made as the Lab project for CSE 4622: Machine Learning Lab.

【Join our constellation of stargazers!⭐️】An interactive AI-powered story generator that creates dynamic narratives through collaborative storytelling between users and Meta's Llama model. Built with Gradio for an intuitive UI and leveraging the Hugging Face Inference API.

A game where you need to guess whether a tweet comes from a human, or from a neural network language model trained on a category of tweets.

Serving large language model with transformers

AI voice assistant made with Streamlit python and powered by Gemini, Mistral and PHI-3. This is a virtual assistant application built in Python that can understand voice commands and complete tasks like opening websites, playing music, telling the time, sending emails, searching Wikipedia, and more.

Elixir RAG library with multi-LLM routing (Gemini, Claude, OpenAI, Ollama), GraphRAG, knowledge graphs, modular retrievers/rerankers, composable pipelines, pgvector integration, advanced chunking, and tool-using agents. Fork of bitcrowd/rag.

A dataset for multimodal machine translation

This repository provides the official resources for EMNLP 2025 Paper Grounding Multilingual Multimodal LLMs With Cultural Knowledge

FeelingBlue: A Corpus for Understanding the Emotional Connotation of Color in Context, accepted at TACL 2022, presented at ACL 2023

building various transformer model architectures and its modules from scratch.

Incorporate Image, Text and Tabular Data with HuggingFace Transformers

LLM, Fine Tuning, Llama 2, Gemma, Mixtral, vLLM, LangChain, RAG, ChromaDB, FAISS

Comprehensive LLMs repo, where I cover both theoretical and practical aspects of LLMs.

Bug detection and fixing using prompt engineering and fine-tuning with Llama 3 and a custom mutation-based dataset. Includes a user-friendly interface and performance evaluation metrics.