IM integration/middleware development for LLM-powered educational chatbot

Abstract

This research explores the development of Elunify, a streamlined platform addressing critical challenges in chatbot deployment and integration with instant messaging (IM) applications. Despite the growing adoption of Large Language Models in various industries, technical complexities in deployment and integration often impede successful implementation, with approximately 30% of AI projects stalling at the proof-of-concept phase. Through an exploratory approach, this study identifies key pain points in existing deployment architectures, constructs a middleware proof-of-concept that abstracts deployment complexities, and evaluates its effectiveness through practical testing scenarios. Focusing specifically on Python-based chatbot backends and Telegram integration, Elunify aims to bridge significant knowledge gaps for both junior developers and experienced chatbot developers with limited deployment expertise. By unifying deployment and IM integration processes, this research investigates how streamlined deployment can facilitate clearer business value demonstration and enhance prototype testing methodologies, potentially offering a more accessible, cost-efficient alternative to existing enterprise-grade chatbot platforms and complex open-source solutions.