Nayan Pal

The system continuously monitors facial landmarks and eye-related patterns through a camera to identify signs of prolonged eye closure and driver fatigue. When drowsiness is detected, it triggers an immediate warning through an onboard buzzer and can interact with vehicle-control hardware to improve safety. The system also integrates GPS and communication capabilities for location tracking and emergency-response scenarios. A Streamlit dashboard provides a real-time interface for monitoring driver status and system activity.

Instead of relying solely on an LLM's internal knowledge, the system retrieves relevant information from trusted medical and pharmaceutical sources before generating an answer. It can analyze user queries, retrieve supporting evidence, classify potential risks, identify drug interactions and provide citations so users can understand where the information comes from. The architecture includes a clarification engine for ambiguous queries, multilingual support, and a drug-interaction knowledge graph.

The system analyzes train, time and compartment-level information to generate safety and crowd-risk scores using machine-learning models and historical patterns. Passenger feedback and environmental factors such as weather can be incorporated to dynamically adjust risk assessments. The platform also includes a multilingual Jan Suraksha Bot for FIR and grievance assistance, plus a police-facing safety analytics dashboard for station hotspots.

The system uses natural-language processing and large language models to analyze legal documents, identify important clauses and surface potentially significant or risky sections. Instead of requiring users to manually inspect lengthy contracts, the platform extracts relevant information and presents it in a structured and accessible form, helping users understand contractual obligations, noteworthy clauses and legal context.