Degree

Master of Science in Computer Science

Department

Department of Computer Science

School

School of Mathematics and Computer Science (SMCS)

Date of Submission

Fall 2025

Supervisor

Dr. Muhammad Saeed, Visiting Faculty, Department of Computer Science

Committee Member 1

Dr. Muhammad Saeed

Keywords

Autonomous AI Agents, DevOps Automation, Enterprise Collaboration Platforms, Intelligent Decision Making, Cloud Infrastructure Management

Abstract

An autonomous AI agent designed to automate DevOps operations in enterprises, by automating rigorous manual tasks done by the DevOps engineers. The project addressed inefficiencies in the current DevOps workflows where continuous manual and human interventions cause delays in operations and increase the risk of errors made by humans. The solution integrates natural language processing capabilities with Slack conversational tools with the AWS infrastructure, and a Modular Control Plane (MCP) server to enable conversational DevOps automation in enterprise environments. Throughout the project lifecycle, we have successfully completed the requirement analysis, established a development environment, implemented a functional interface, for interactions with AWS environment, and achieved NLP integration with intent recognition and entity extraction from the user’s prompt. We did extensive end-to-end testing of the system for reliable communication between the interface currently having a slack bot and AWS services through the API / AWS SDK layer, with successful execution of both simple and complex commands executed by DevOps in their daily responsibilities. We tested the systems automated DevOps tasks such as EC2 instance creation, updation, S3 operations like listing bucket information and files, CloudWatch monitoring, and some infrastructure scaling commands, reducing average response time from 3-5 minutes executed by humans subject to their availability to around 10 to 15 seconds using slack bot. We also did security testing and checked authentication, authorization flows, and compliance with enterprise security standards. This research contributes to the emerging field of autonomous AI systems in enterprise environments by demonstrating deployment of intelligent agents that are capable of independent executions of commands and decision-making via LLMs in critical infrastructure management.

Document Type

Restricted Access

Submission Type

Research Project

Download

The full text of this document is only accessible to authorized users.

Share

COinS