What Are AI Agents? A Short Intro And A Step-by-Step Guide to Build Your Own.

by

Wow.

So, what is really behind the trend? The key to understanding agents is agency.

Unlike traditional generative AI systems, agents don’t just respond to user input. Instead, they can process a complex problem such as an insurance claim from start to finish. This includes understanding the text, images and PDFs of the claim, retrieving information from the customer database, comparing the case with the insurance terms and conditions, asking the customer questions and waiting for their response — even if it takes days — without losing context.

The agents do this autonomously — without humans having to check whether the AI is processing everything correctly.

The Espresso Machine and the Barista

In contrast to existing AI systems and all the copilots out there that help employees to do their job, AI agents are, in fact, fully-fledged employees themselves, offering immense potential for process automation.

Imagine — an AI that can take on complex, multi-step tasks that are currently performed by a human employee or an entire department:

  • Planning, designing, executing, measuring, and optimizing a marketing campaign
  • Locate a lost shipment in logistics by communicating with carriers, customers, and warehouses — or, if it remains lost, claim its value from the responsible partner.
  • Search the trademark database each day and determine whether a new trademark has been registered that conflicts with my own trademark and immediately file an opposition
  • gather the relevant data or ask employees, check the data and compile an ESG report

Currently, AI models can assist with tasks like generating campaign content or evaluating emails, but they lack the ability to execute an entire process. An AI agent can do that.

Traditional generative AI can help human teams in a process (yellow), AI agents can execute the complete process end2end (orange). Image credit: Maximilian Vogel

While traditional models are like great espresso machines, agent-based AI is the barista. Not only can they make coffee, but they can welcome the guests, take the order, serve the coffee, collect the money, put the cups in the dishwasher, and even close up shop at night. Even the best espresso machine in the world can’t run a café by itself, but the barista can.

Why can the AI agent and the barista do this? They excel at mastering various subprocesses of a complex job and can independently decide which task to tackle next. They can communicate with people, like the clients, if they need more information (milk or oat milk?). They can decide who they should ask in case of problems (beans are out => boss, coffee machine is on strike => customer service of the machine vendor).

AI agents vs. traditional generative AI. Image credit: Maximilian Vogel

Anatomy of an AI Worker

But enough chatting, let’s build an AI agent. Let us have a look at the relevant processes and workflows.

Let us build an agent for the insurance process shown in the diagram above. The agent should handle an insurance claim from start to reimbursement.

What we are developing here is the business architecture and the process flow. Unfortunately, I can’t dive into the coding because it can quickly become very extensive.

1. Classification & sending a job into processing lanes

Our workflow starts, when a customer sends a message with a claim for their home insurance to the insurer.

What does our agent do? It determines what the customer wants by analyzing the message’s content.

Based on this classification, the system initiates a processing lane. Often, this goes beyond function calling; it involves making a fundamental decision about the process, followed by executing many discrete steps.

Leave a Comment