Student Name

Shaheryar AhmedFollow

Degree

Master of Science in Data Science

Department

Department of Computer Science

Faculty/ School

School of Mathematics and Computer Science (SMCS)

Date of Submission

Fall 2024

Supervisor

Dr. Muhammad Sarim, Visiting Faculty, Department of Computer Science, School of Mathematics and Computer Science (SMCS)

Keywords

Agentic AI, Machine Learning, LLM

Abstract

The space of Generative AI has started seeing a shift towards AI Agents. It’s basically an emergence of compound AI systems that can work to solve more complex problems. To tackle such tasks, we need a system that behaves more like a human i.e taking a multi-step approach and breaking the task down into smaller problems. AI agents represent the forefront of Generative AI, advancing beyond basic query-response systems to become collaborative entities. By incorporating capabilities like reasoning, action-taking, feedback processing, and memory, these agents transcend mere tools, evolving into collaborative partners across various industries, with the ability to interact and cooperate to address complex challenges. The purpose of this study is to develop and explore the usability of AI agents in the travel industry. AI agents are advanced LLM systems designed to perform complex tasks. They have the ability to plan ahead, recall previous interactions, and utilize various tools to tailor their responses according to the context and desired style. The methodology involves building customized agents as per requirements e.g in our case it involves setting up multiple agents; one for planning itinerary, one for planning accommodations etc. Building agents means customizing each agent with a specified persona. This means setting it up with specific characteristics that makes it better suited for specific tasks and instructions. These individual agents use the tool they are equipped with to search and complete their respective tasks.

Document Type

Restricted Access

Submission Type

Research Project

Loading...

Media is loading
 

Available for download on Monday, June 16, 2025

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

Share

COinS