Coding agents are like having a personal assistant (think Jarvis). Unlike a standard chatbot that just gives you advice, an agent takes your request, executes the necessary steps, and completes the task while you simply describe what you need. Operating from your terminal, these tools can read files, write code, and fix their own errors until the job is done.

This series is for anyone held back by the technical side of building. Whether you are:

  • A student visualizing data for a project
  • A researcher summarizing a mountain of PDFs
  • Someone wanting to automate a repetitive task

You don’t need to be an engineer. Agents handle the “how” so you can focus entirely on the “what.” Best of all, it’s accessible: open-source options like OpenCode work with Gemini’s free tier, while tools like Claude Code offer advanced features for a fee.

A tour of what’s possible

I asked the agent to build a functional photo editor from scratch.

The agent selected gradio and pillow, installed them, and wrote the interface code. Within minutes, I had a working web app for resizing, rotating, and applying artistic filters. I didn’t write a single line of code; I just described the features.

Next, I tested its ability to handle “heavy reading” by giving it a complex economic report to turn into a presentation.

The agent identified key data points—like oil price shifts and GDP impacts—and designed an eight-slide deck in both PowerPoint and PDF formats. This is a massive time-saver for anyone distilling long documents.

Finally, I gave the agent a folder of messy raw data and asked for an interactive dashboard and a matching slide deck.

Using pandas, the agent cleaned the files and built a dashboard with clickable charts. Since it used the same data for both the dashboard and the slides, everything matched perfectly—acting as both a data analyst and a graphic designer.

Our roadmap

  1. The Fundamentals: Terminal basics, Python, and how agents think.
  2. Setting Up Your First Agent: Getting started with OpenCode and Claude Code.
  3. Customizing the Experience: Using custom prompts and adding new skills.
  4. Sharing Your Work: Putting your creations online for others to use.

The next part covers technical foundations. Knowing a little about the terminal or how AI makes decisions helps when building complex projects. Once we have the basics down, the real fun begins.