⚡️ Local RAG API using FastAPI + LangChain + Ollama | Upload PDFs, DOCX, CSVs, XLSX and ask questions using your own documents — fully offline!

Mobile-first frontend for TaskWizard — a smart AI-powered task breakdown planner. Built with Expo + React Native, connecting to a FastAPI + Ollama backend to generate actionable step-by-step roadmaps from any goal.

Multi-model AI assistant built using Gemini API and Streamlit.

REFRAME helps you rewrite workplace feedback and everyday messages with the right tone — empathetic, constructive, or persuasive — powered by free LLMs via OpenRouter.

A type-safe graph execution framework built on top of OpenLit for LLM pipelines

A Retrieval-Augmented Generation (RAG) chatbot using OpenAI GPT-3.5 and Streamlit.

This repo builds a Streamlit-based app that converts natural language queries into SQL using Gemini Pro, enabling real-time interaction with a live SQL database.

A no-code AI-powered writing assistant that helps users generate creative stories, poems, and essays — built using HuggingChat and hosted on Hugging Face.

A lightweight, modular Retrieval-Augmented Generation (RAG) system built with Streamlit, FAISS, and LLMs like OpenAI and Ollama. Upload documents, embed them, and ask intelligent questions with real-time context-aware responses.

A conversational Form-Filling Agent built using LangGraph and LangChain. It interactively collects user information such as name, age, and email. The agent manages conversation state across turns and intelligently prompts the user for any missing details in a natural, chatbot-style flow.