AI Agents Explained: How Autonomous AI Agents Work

What is an AI Agent?

A standard LLM call is stateless — you send a prompt, get a response, done. An AI agent is different: it's an LLM that can take actions, observe results, and decide what to do next in a loop until it completes a goal.

Think of the difference between asking someone "What's the weather in Tokyo?" (single LLM call) versus asking "Book me the cheapest flight to Tokyo next week" — the second requires multiple steps: searching flights, comparing prices, checking availability, and making a booking. That's an agent.

The ReAct Loop

Most agents follow the ReAct pattern (Reason + Act), alternating between thinking and doing:

This loop continues until the agent decides it has enough information to give a final answer, or hits a maximum iteration limit.

Agent Components

Building a Simple Agent

import OpenAI from 'openai';

const openai = new OpenAI();

// Define tools the agent can use
const tools = [
  {
    type: 'function',
    function: {
      name: 'calculate',
      description: 'Perform a mathematical calculation',
      parameters: {
        type: 'object',
        properties: {
          expression: { type: 'string', description: 'Math expression to evaluate' }
        },
        required: ['expression'],
      },
    },
  },
];

// Tool implementations
function executeTool(name: string, args: Record<string, string>) {
  if (name === 'calculate') {
    return String(eval(args.expression)); // Use a safe math parser in production!
  }
  return 'Tool not found';
}

// Agent loop
async function runAgent(userMessage: string) {
  const messages: OpenAI.Chat.ChatCompletionMessageParam[] = [
    { role: 'system', content: 'You are a helpful assistant with calculation abilities.' },
    { role: 'user', content: userMessage },
  ];

  for (let i = 0; i < 10; i++) { // Max 10 iterations
    const response = await openai.chat.completions.create({
      model: 'gpt-4o-mini',
      messages,
      tools: tools as OpenAI.Chat.ChatCompletionTool[],
      tool_choice: 'auto',
    });

    const message = response.choices[0].message;
    messages.push(message);

    if (response.choices[0].finish_reason === 'stop') {
      return message.content; // Agent is done
    }

    // Execute tool calls
    for (const toolCall of message.tool_calls ?? []) {
      const args = JSON.parse(toolCall.function.arguments);
      const result = executeTool(toolCall.function.name, args);
      messages.push({
        role: 'tool',
        tool_call_id: toolCall.id,
        content: result,
      });
    }
  }
  return 'Max iterations reached';
}

const answer = await runAgent('What is 15% of 847, plus the square root of 144?');
console.log(answer);

Multi-Agent Systems

Complex tasks benefit from multiple specialized agents working together:

Agent Frameworks