Asana completed five years of engineering work in just two weeks using OpenAI Codex. The company needed to remove Enzyme, an outdated testing system that had become a major obstacle to modernizing its frontend stack. Enzyme had fallen out of active maintenance, making code upgrades increasingly difficult.

Asana used Codex, powered by frontier models, to tackle the large codebase changes. From a five-sentence prompt, up to four coding agents worked in parallel across separate copies of the codebase. An engineer checked progress twice daily and reviewed every proposed change. Simpler instructions proved more effective than complex setups.

After 1.5 weeks of engineering effort across two calendar weeks, Enzyme was fully removed. Model and infrastructure costs totaled approximately $12,000. For comparison, the previous staffing plan was estimated to take at least five years and cost roughly $6 million.

The experience changed how Asana views long-running software projects. With this migration complete, the company can now test AI agents on other migrations, rewrites, and performance problems once assumed to require years of effort.