The Evolution of Software Development

Claude code and Opencode

Let’s start with a chatbot. You can have a conversation with it, and it will provide answers that are accurate—or at least approximately correct. So when you’re coding, you might decide to ask the chatbot for help. You copy your code, show it to the chatbot, and then use the code it generates. You might think that constantly copying and pasting code is tedious. So why not create a tool that can directly read all your code—while handling context limits—and modify it for you? This is exactly how Claude Code was born. This can improve coding efficiency by 5 to 10 times.

As for me, I choose to use OpenCode. Nowadays, coding agents have become an essential tool for software engineers. I want to use my own tool, whether I’m at work or coding at home.

From the traditional software to really world

As people’s coding skills improve, I believe—or hope—that the software they create will become more powerful as well. This opens up a whole new way of thinking about the future of software. When you talk to a chatbot, how much do you think it feels like a real person? And in the future, how human-like could it become? For example, I was amazed by news that Claude Code was given some functions to interact with a physical environment—like controlling a seedling. The results were surprisingly effective. Like in the example, robots and machines can all be endowed with this kind of intelligence (I think we can now call chatbots a form of intelligence). In the future, every machine could have its own dedicated software, effectively giving each device its own intelligence.

Another Way to Think About It

I’m still a bit unsure about which area I should focus on.

But I think there’s another way to approach this problem—by looking at the form of the work itself. For example, I would choose projects that are used by many people and collaborate with people I truly admire.