Introduction

The famous AI book Artificial Intelligence : A Modern Approach cites agent as follows :

An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators

Russell & Norvig, AI: A Modern Approach (Fourth Edition) - Chapter 2.1

Agents interact with their environment and typically consist of several important components:

  • Environments — The world the agent interacts with
  • Sensors — Used to observe the environment
  • Actuators — Tools used to interact with the environment
  • Effectors — The “brain” or rules deciding how to go from observations to actions

agent_def.png

This framework is used for all kinds of agents that interact with all kinds of environments, like robots interacting with their physical environment or AI agents interacting with software.

The AI Agent

With recent development in the field of AI and the popularisation of the LLMs leads to a new agent architecture, Meta AI research coined it as The Augmented LLM

The base idea is through external systems, the capabilities of the LLM can be enhanced. the external system generally consists of the tools and memory.

augmented_llm.png

For instance, when faced with a math question, the LLM may decide to use the appropriate tool (a calculator).

calc_example.png

So is this “Augmented LLM” then an Agent? No, and maybe a bit yes.

An augmented LLM is not traditionally considered an agent because it lacks autonomous planning and goal-directed behavior. However, as it is granted persistent memory and the autonomy to dynamically select tools, evaluate outcomes, and determine its own sequence of actions in pursuit of a goal, it evolves into an agentic system, making the distinction increasingly blurred.

We can generalize the agent definition and framework that we studied earlier, a bit to make it suitable for the “Augmented LLM”

augmented_llm_agent.png

Using the “Augmented” LLM, the Agent can observe the environment through textual input (as LLMs are generally textual models) and perform certain actions through its use of tools (like searching the web).

To select which actions to take, the LLM Agent has a vital component: its ability to plan. For this, LLMs need to be able to “reason” and “think” through methods like chain-of-thought, using this reasoning behavior, LLM Agents will plan out the necessary actions to take.

reasoning_ai_agents.png

This planning behavior allows the Agent to understand the situation (LLM), plan next steps (planning), take actions (tools), and keep track of the taken actions (memory).

augmented_llm_with_specs.png

Depending on the system, you can LLM Agents with varying degrees of autonomy, Depending on who you ask, a system is more “agentic” the more the LLM decides how the system can behave.

ai_agent_control.png

When to build AI Agents

Before we move forward it is important to know that when to build the AI Agents, the following tasks or problems can be resolved by an AI Agent:

Category Description Example
01 Complex decision-making Workflows involving nuanced judgment, exceptions, or context-sensitive decisions. Refund approval in customer service workflows.
02 Difficult-to-maintain rules Systems that have become unwieldy due to extensive and intricate rulesets, making updates costly or error-prone. Performing vendor security reviews.
03 Heavy reliance on unstructured data Scenarios that involve interpreting natural language, extracting meaning from documents, or interacting with users conversationally. Processing a home insurance claim.

If your usecase don't follow any of these catagory then you may need to re-think your decision of building an AI Agent because then it is higher probability that other simple and a deterministic solution may be sufficient for the use-case.

Components of AI Agents

Fundamentally AI Agents consists of the few components :

Component Description
01 Model The LLM powering the agent’s reasoning and decision-making
02 Tools External functions or APIs the agent can use to take action
03 Instructions Explicit guidelines and guardrails defining how the agent behaves
04 Memory Mechanisms that allow the agent to store, recall, and build upon past interactions and context
05 Planning / Reasoning The cognitive framework that enables the agent to break down complex objectives, handle multi-step problems, and adapt to changing conditions.

We'll discuss all the components in detailed in upcoming articles.