Only About 1% of the Tokens Thariq Generates Go Into Production Code

Why It Matters

Lower generation costs change engineering and product workflows by encouraging rapid experimentation, lightweight documentation and emphasis on plan quality over format, which could speed decision-making and reduce friction around spec ownership. That shift may reshape how teams structure processes, tools and responsibilities for delivering production code.

Summary

The speaker says only about 1% of the tokens they generate end up in production code, with the vast majority used for dashboards, custom interfaces and exploratory work. As generation costs approach zero, rigid rules about single-source templates for specs and PRDs become unnecessary, enabling documentation to be created in whatever format is most useful. This allows teams to focus on the substance—whether a plan is a good idea and can be executed—while producing high-quality, disposable “just-in-time” documentation and throwaway software. The speaker also emphasizes a collaborative workflow with AI (Claude) that yields more polished, personalized outputs.

Original Description

When tokens are cheap, you can afford to make everything you interact with beautiful and tailored to your specific needs. The hope is that this richness in the process translates into better final products.