What's your experience with Agentic Coding?

Question

I have been experimenting more deeply with agentic coding, and it’s made me rethink how I approach building software. One key difference I have noticed is the upfront cost. With agentic coding, I felt a higher upfront cost: I have to think architecture, constraints, and success criteria before the model even starts generating code. I have to externalize the mental model I normally keep in my head so the AI can operate with it. What I have realized is that agentic coding shifts my cognitive load from on-demand execution to more pre-planned execution. My role is less about 'precisely' implementing every piece of logic and more about defining the problem space clearly enough that the agent can assemble the solution reliably.

HackerNews · Accepted Answer

Agentic coding is changing the way developers approach software development by shifting the focus from writing precise code to defining the problem space and letting AI models generate the solution. This approach requires a higher upfront cost in terms of planning and defining the architecture, constraints, and success criteria. However, it also enables developers to work more efficiently and effectively by delegating coding tasks to AI agents. As a result, developers need to take on a more QA-oriented role to evaluate the output of the AI models.

What's your experience with Agentic Coding?

Resources