Agentic loops excel in constrained tasks with clear feedback, like fixing code based on an AI-generated review score. They fail in open-ended creative tasks like building an application, where they make costly, incorrect assumptions about product details.
The primary barrier to using agentic loops is cost. They consume vast amounts of tokens making assumptions, a luxury only affordable to well-funded AI researchers, not the average developer on a budget plan. For most, it's an unproductive way to burn money.
The idea of an AI building an app from a single spec file is flawed because no document can capture every product detail, edge case, or evolving requirement. This forces the AI to make assumptions, which are almost always misaligned with the creator's vision.
AI loops and tools like `/goal` are effective for quickly building experimental prototypes where fine details are unimportant. For building a polished product where details and unique "sauce" matter, the human-in-the-loop approach remains superior and more cost-effective.
The most effective method for building apps with AI is still the iterative "human-in-the-loop" process. A human directs the AI with prompts, reviews the output, and provides corrections. This allows for creative control and avoids the costly, assumption-driven errors of fully autonomous loops.
Agentic loops are suitable for tasks where the output is binary (done or not done) and creativity is not required. Generating hundreds of SEO pages from a fixed template is a prime example where automation excels, unlike building a unique user-facing application.