P

python

A Flask budget trip planner for affordable destination, hotel, transport, and itinerary recommendations.

An offline-first, CPU-optimized suite of classroom helper tools for teachers and students in low-connectivity areas. Powered by local OCR (PaddleOCR) and local Small Language Models (Qwen 2.5 via Ollama) to automate grading and flashcard generation entirely on a standard laptop

OriginAI transforms research papers into startup opportunities, investor reports and pitch decks using AI.

CivicEye AI is a civic issue reporting and citizen engagement platform that helps users report public problems through AI-assisted workflows, location tracking, and community-driven issue monitoring.

Offline CPU-first AI application that converts PDFs, images, audio, and text into structured study notes using local AI models.

WardroAI is an AI-powered wardrobe assistant that analyzes clothing items from images, categorizes them, stores them in a local catalog, and recommends outfit combinations—all while running completely offline for privacy and speed.

Offline-first lecture transcription and structured study data extraction CLI/PWA.

FormForge AI is a smart form processing application that uses OCR and artificial intelligence to extract data from printed and handwritten forms and convert it into structured digital records for easier storage, search, and export.

Resume-Analyzer is an offline, CPU-first AI application that analyzes resumes against job descriptions and converts unstructured resume data into structured insights. It extracts skills, experience, identifies missing requirements, calculates a job match score, and provides improvement recommendations—all without requiring an internet connection or cloud APIs.

An AI-powered bilingual assistant helping citizens discover government welfare schemes they're eligible for and draft RTI requests. Uses RAG with Google Generative AI, voice input, and real-time eligibility checking for Telangana and India-wide schemes.

Offline-first tribal health app

Smart Cab Finder is an AI-powered ride comparison platform that helps users compare fares, estimated arrival times, and ride options across multiple cab providers. The application includes an intelligent AI assistant for ride recommendations, supports Bring Your Own Key (BYOK) integrations, and is designed to support local AI inference through Ollama. To improve accessibility, the platform provides multilingual support in English, Hindi, and Telugu. Built using Python and Flask, the project follows secure development practices with automated testing, CI/CD pipelines, security scanning, and compliance-driven documentation.

Offline, CPU-only document to structured JSON extractor

Offline-first meeting transcriber using whisper.cpp and Qwen2.5 — fully local CPU inference, no cloud APIs, no GPU required.

Offline Medical Prescription Analyzer – Privacy-first offline OCR system using Tesseract for prescription extraction, validation, and CI/CD compliance automation.

Offline-first meeting intelligence powered by local AI.

Harry Potter RAG Chatbot is a Streamlit-based conversational app that answers Harry Potter questions by retrieving relevant text from a custom dataset using semantic embeddings and FAISS, then generating grounded responses via the Groq LLM.

SafeRoute Guardian AI is a Python Streamlit web application that helps users plan safer travel routes and stay connected with trusted guardians. It combines route planning, safety scoring, location-based risk awareness, AI-assisted guidance, and notification workflows into a single interactive dashboard.

The app runs from streamlit_app.py and uses helper logic in app/streamlit_helpers.py, with SQLAlchemy models and service modules handling database access, route safety, AI responses, notifications, and guardian features. For local development, it is configured to use SQLite through a root .env file, making it easy to run as a demo or hackathon project.

In short: SafeRoute Guardian AI is a safety-focused navigation and guardian support tool designed to make travel decisions more informed, proactive, and secure.

Lifeline AI is an AI-powered emergency response platform providing real-time emergency intelligence, ambulance coordination, hospital connectivity, SOS management, and disaster response monitoring.

Project Overview: The CPU-First EMS is a highly resilient web platform designed specifically for distributed teams and field workers operating in environments with unreliable internet access.

Key Technical Achievements:

Offline-First Reliability: Implemented robust client-side storage to queue submissions and track attendance locally, ensuring zero data loss during outages and enabling automatic synchronization when reconnected.

Architectural Overhaul: Successfully migrated the application from a local Streamlit script to a scalable, production-ready REST API using Flask and Gunicorn with multi-threaded workers.

Modern User Experience: Built a responsive, dynamic frontend using Vanilla JavaScript and the Fetch API, styled with a modern glassmorphism aesthetic.

Production-Ready Deployment: Engineered a secure, multi-stage, non-root Docker build to streamline deployment and ensure consistent environments across local and production servers.