We continue our series of articles on AI agents. In the previous article, our author from Tajikistan, Muhammad Abdugafarov, shared the Copilot configuration methods he uses personally, and in this article, he will explore the possibilities of creating AI agents and why this direction opens up new prospects for developers.

Muhammad Abdugafarov, city — Khujand, team lead, programmer at Silk Road Professionals, LinkedIn

Why AI Agents Are a Window of Opportunity

AI agents are a new area in development where the market of specialists is still small, but interest from companies is growing. For beginners, this is a chance to enter IT and find their niche, as the demand for agent developers and AI solutions is constantly increasing.

How AI Agents Complement Classic Applications

At the moment, agents do not completely replace backend services but add a new level of interaction with the user. For example, if previously the user manually searched for the required information through filters and websites, now the agent can analyze the data itself and provide a relevant answer. An example is the classic Google search, where the user searches for what they need themselves, and Perplexity AI, where the agent analyzes websites itself and immediately provides the final answer.

What You Need to Know to Get Started

To create agents, basic knowledge of a programming language — most often Python or JavaScript — and the basics of object-oriented programming are sufficient. There are many frameworks and ready-made components that allow you to quickly create agents. The complexity lies not so much in writing code but in creating a truly useful agent that solves the user's tasks.

Example: An Agent for Searching Direct Flights

An agent is, in essence, a class to which you assign a name, instructions, and the necessary tools for interaction. For example, you can create an agent that uses the google_search tool to find a direct flight between two specified cities. The user enters their query, and the agent automatically forms a search query, analyzes the results, and provides the most relevant direct flight option, responding in text form. Such an agent saves the user's time and simplifies the process of searching for airline tickets.

Agent Code Example on ADK Using Gemini

You can get a demo project with a ready-made agent in this GitHub repository and test its operation yourself.

By running this agent, you can test its operation. Below is an example of a conversation:

Multi-Agent Systems: Examples

You can create more complex systems where agents work in parallel or sequentially. For example, when writing an article:

— the first agent collects ideas
— the second chooses the headline
— the third writes the text
— the fourth checks facts and adds links.

The result is a higher-quality material due to the joint work of agents.

What Other Frameworks to Use

Besides ADK, there are other popular tools: LangChain, Llama, and others. They allow you to create agents for various tasks and quickly prototype solutions.

Conclusion

Creating agents is a new direction that is changing human interaction with programs. I recommend everyone looking for new opportunities in programming to study and practice agent technologies. The future is already here — you can create your own agents and take the user experience to a new level.