I Don't Need Smart AI to Build Apps. Here's Why.
I stopped reaching for the most capable model for every app. With TDK CLI, the infrastructure is already decided, so a simpler agent only has to get the application logic right. For a long time I assumed that building an app with AI needed the best model available. Every project got the most capable assistant, the longest prompts, and the most careful review. Most of the effort went into getting…
The author of the article no longer feels the need to use the most advanced AI models to build applications. Instead, they advocate for using simpler agents with TDK CLI, which streamlines the infrastructure setup and reduces the need for complex prompts. The article explains that TDK CLI handles most of the infrastructure decisions, leaving the agent to focus on the core application logic.
The author provides three prompts for using a simpler agent with TDK CLI:
1. Building a frontend that calls an API: The agent creates a React frontend that fetches data from an API client generated by TDK CLI. The agent only changes one file, uses the provided client, and verifies the result using a dry-run command.
2. Creating a backend with a database: The agent generates a Hono backend with a database and a GET route returning a JSON array. The agent leaves the infrastructure unchanged and uses a dry-run to verify the services' starting order.
3. Setting up a worker that reads a queue: The agent creates a worker that logs "processing" every five seconds and depends on another service. The agent adds the dependency and verifies the result using a dry-run command.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
