S

swecha

Offline-first, CPU-first AI reader companion that narrates and analyzes stories locally using Ollama and structured feedback JSON.

This is a journaling website made by me, all the related files are in this repo

HealthCamp OS: A multi-tenant platform for NGOs and doctors to manage health camps. Features Gemini AI triage, volunteer management, and responsive impact reporting.

NyayaBloom is an AI-powered Judicial Intelligence Platform designed to democratize legal knowledge and improve judicial transparency across India. The platform enables citizens, lawyers, law students, researchers, journalists, and judicial professionals to access legal information, understand judicial processes, explore case precedents, and improve legal literacy through a multilingual AI-powered ecosystem. Instead of replacing judges, courts, or legal professionals, NyayaBloom serves as an intelligent legal knowledge assistant that simplifies legal information and promotes informed civic participation.

StoryMirror CLI is a CPU-first command-line application that transforms unstructured stories into meaningful insights through local AI processing. It provides story analysis, multilingual text-to-speech (TTS), runtime performance metrics, and customizable terminal themes, all while running on a local FastAPI backend.

Designed for offline and resource-efficient environments, StoryMirror performs inference on the CPU without requiring a dedicated GPU, making it suitable for developers, students, and writers who need an accessible AI-powered storytelling tool.

Key Features 📖 Analyze stories from text or files 🎙️ Generate multilingual text-to-speech narration 🎨 Customize the CLI with multiple themes 📊 Monitor CPU usage, memory, and inference performance ⚡ Run entirely on a local FastAPI backend 💻 CPU-first design with no GPU dependency 🌐 Supports offline or local deployment 🚀 Simple command-line interface for quick workflows

MeetingMind is an offline-first, CPU-first AI meeting assistant that turns raw meeting transcripts into structured summaries (agenda, attendees, decisions, action items, deadlines, open questions, notes), built with React 19 + Vite + TypeScript on the frontend and Python Flask + SQLite on the backend. It runs entirely locally without cloud LLM APIs, using a modular AIAdapter that currently mocks responses but is designed for drop-in llama.cpp integration (for summarization) and whisper.cpp (for offline audio transcription), so it works on consumer CPUs with full data privacy and no internet dependency.

StreetCheck is an application for road and civic safety, a per-road-segment score based entirely on public verifiable,non-sensitive data for the swecha hackathon

Deepdive AI is a research assistant and makes pdfs and powerpoint presentation for a given topic.

AI-powered open-source repository onboarding assistant that analyzes GitHub repositories and provides onboarding guidance using Gemini and Ollama.

newhackathon-qlhmvw7jv34agekacwf76n.streamlit.app

The Image Captioning project is an AI-based system that automatically generates textual descriptions for images. It combines computer vision and natural language processing (NLP) techniques to understand the content of an image and produce a human-readable caption.

The system uses deep learning models, typically a Convolutional Neural Network (CNN) for extracting image features and a Recurrent Neural Network (RNN) or Transformer-based model for generating sentences. This enables applications such as assisting visually impaired users, enhancing image search engines, and automating social media content tagging.