From Search Algorithms to Agentic AI: What Recent AI Research Taught Me

24P-0683 Zarish Imran — Sat, 14 Mar 2026 17:36:57 +0000

Introduction

Artificial Intelligence is not only about building models but also about designing intelligent systems that can search, plan, and make decisions in complex environments.To understand how these ideas are applied in real research, I explored two recent AI papers:

The Rise of Agentic AI: A Review of Definitions, Frameworks, and Challenges (2025)

Research on the A Algorithm Based on Adaptive Weights (2025)*
Paper 1: The Rise of Agentic AI

*Goal of the Paper
*
The main goal of this paper is to explain the concept of Agentic AI, which refers to AI systems that can plan tasks, make decisions independently, and interact with their environment.

Connection to Our AI Course

This research connects directly with the Intelligent Agents topic we studied in class.

In our course we learned about:

Simple Reflex Agents
Model-Based Agents
Goal-Based Agents
Utility-Based Agents

For example, a rescue robot in a disaster environment could behave like an agentic AI system that:

searches for survivors
plans safe paths

adjusts its decisions based on new information

*Paper 2: Adaptive A Search Algorithm

Goal of the Paper**

The second paper focuses on improving the A* search algorithm, which is widely used for pathfinding and optimization problems.The traditional A* algorithm uses a heuristic function to estimate the cost from the current node to the goal. However, the accuracy of the heuristic greatly affects performance.

Connection to Our AI Course

In our course we studied several search algorithms including:

Breadth-First Search (BFS)
Depth-First Search (DFS)
Uniform Cost Search (UCS)
A* Search

The improved A* algorithm discussed in the paper shows how researchers continue to refine classical algorithms to solve real-world problems more efficiently.

*Personal Insights (Manual Reading vs NotebookLM)
*
Reading research papers manually helped me understand the motivation behind the research, such as why existing algorithms are not always sufficient in complex environments.

What I Found Most Interesting

The most interesting insight for me was how traditional AI concepts from our course are still the foundation of modern research.

*For example:
*

Agent architectures are evolving into Agentic AI systems capable of autonomous decision making.

Conclusion

Exploring recent AI research helped me see how foundational concepts like agents and search algorithms continue to evolve in modern AI systems.

Special thanks to Raqeeb eebr for the assignment guidance.

BLOG LINK:

https://youtube.com/shorts/sy9OeKruP5o?si=ExolScs8i1SOmxUm

DEV Community: 24P-0683 Zarish Imran

From Search Algorithms to Agentic AI: What Recent AI Research Taught Me