Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{183238,
        author = {Dr. Chiranjeevi Kommula and Narayana Dasari and Murahari Satish Kumar},
        title = {AGENTIC AI: EVOLUTION, IMPLEMENTATION DOMAINS, COMMUNICATION PROTOCOLS AND PRACTICAL USE CASES},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {3},
        pages = {889-905},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=183238},
        abstract = {This is signified by the development of Agentic Artificial intelligence (AI) which introduces a revolutionary change towards the operation of the intelligent systems; that is, it shifts toward the use of autonomous agents who possesses the abilities to engage in goal-driven reasoning, learning, and interaction. This paper discusses the development of Agentic AI and its basis, outlining the defining characteristics of the latter, including the agents’ capabilities of autonomy, memory integration and multi-modal communication. Focusing on the differences between agentic frameworks and the conventional concepts of AI, the study formulates the improvement in concepts and functions the new paradigm introduces. One of the most important features of Agentic AI is the allowing of the Large Language Models (LLMs) to use as the cognitive engines to drive powerful reasoning and decision-making functions. Other enabled technologies that are discovered in the article include Retrieval-Augmented Generation (RAG), vector databases, Kafka-based streaming stacks, and cache-augmented generation among others that play a key role in improving the system performance, memory and flexibility. Simultaneously with that, we examine the layered communication architectures, which enable the smooth integration of coordination and preservation of context within the agents, e.g. Agent-to-Agent (A2A) and Model Context Protocol (MCP). The agentic systems have a concept of memory that is central to the way agents act, learn, and remember in the episodic, semantic, procedural and working memory. Memory is a major component in agentic systems and it is differentiated into episodic, semantic, procedural and working memory, which influences memory learning, recalling and acting by agents. Another area that is covered in the article is the significance that should be attached to tracking, monitoring and transparency on the actions of the agent. Finally, application scenarios in the real world that have been seen in the enterprise automation, healthcare, education and customer service sectors are demonstrated as the practicality and scalability of the Agentic AI systems in both the present and the future applications.},
        keywords = {Agentic AI, Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), Vector Databases, Kafka, Cache-Augmented Generation, Agent Communication Protocols, Model Context Protocol (MCP), Episodic Memory, Semantic Memory, Procedural Memory, Working Memory, Tracing and Monitoring, Autonomous Agents, AI Implementation Domains.},
        month = {August},
        }