MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202621051313 A) filed by Ayush Kumar Choudhary; Priyanshu Kumar Dubey; Saurav Kumar Singh; Saurabh Kumar; Vivek Patil; and Moresh Madhukar Mukhedkar on April 22, 2026, for Abhiyanta Gpt - Ai-Powered Engineering Assistant System.
Inventors include Ayush Kumar Choudhary; Priyanshu Kumar Dubey; Saurav Kumar Singh; Saurabh Kumar; Vivek Patil; and Moresh Madhukar Mukhedkar.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: The present invention relates to an AI-powered engineering assistant system designed to provide intelligent, context-aware, and domain-specific technical assistance through a web-based conversational interface. The system integrates multiple functional layers, including a user interface layer, a session management module, a data persistence mechanism, and an API interaction layer, to enable seamless and efficient communication between users and an external large language model. The user interface layer is designed to provide a responsive and intuitive chat-based interaction environment, supporting multiple device formats including desktop, tablet, and mobile platforms. The system incorporates advanced session management capabilities that allow users to create, access, manage, and delete multiple chat sessions, thereby enabling structured and organized interaction workflows. A persistent data storage mechanism, implemented using browser-based storage, ensures continuity of conversations across sessions without reliance on external databases. The system is specifically optimized for engineering applications and supports queries across multiple domains, including civil, mechanical, electrical, electronics, computer, chemical, and production engineering. It incorporates predefined quick-action prompts to guide users in formulating technical queries and enhance usability, particularly during initial interaction stages. The API interaction layer facilitates communication with an external AI model through a chat completion framework. This layer performs critical operations including prompt sanitization, response optimization, rate-limit handling, and error management. Additionally, the system incorporates fallback mechanisms for model selection and response generation to ensure robustness and uninterrupted service in cases of API failure or limitations. The invention further enhances reliability and user experience through structured response generation, contextual awareness, and efficient handling of conversational state. By integrating domain-specific intelligence, persistent session handling, and adaptive user interface design, the proposed system significantly improves the accuracy, usability, and accessibility of engineering assistance platforms
Disclaimer: Curated by HT Syndication.
