How to prepare for AI-driven code modernization projects

Two Anthropic engineers advise companies modernising old code with AI agents to agree, before the work starts, what each change must prove and how much human review it gets.

In banks and other regulated companies, every change to a critical system is reviewed and approved, a process built for code written by people. When AI agents rewrite an old system, they produce changes far faster than anyone can review them one by one. Jonah Ezekiel and Lexie Tonelli, engineers who work with Anthropic's customers, set out how to prepare such a project, drawing on those deployments.

• Before the work starts, the team agrees on automatic checks that every change must pass, such as the old tests passing and old and new code giving the same output. The authors call this set of checks a certificate.
• The checks are written with the people who will review and approve the changes. A good test of them is whether those people would be comfortable merging a change on their evidence alone.
• A written, tiered review policy, agreed in advance, sets how much human review each change gets according to how much it could break. Critical parts keep full review, and experts spend their time on the riskiest changes.
• The authors advise that this policy come from the top of the organisation. Responsibility for a defect that reaches production is then shared rather than pinned on one approver.

Agreeing on checks and a review policy with the approvers before agents start could suit other agent-built work; this is an inference and is not the authors' claim. The article is a guide from Anthropic, which sells the tools, and it reports no measured results.

Derived by Working Surface from the article. Source line: Write the certificate with the people who will review and promote changes into production.

24 September 2026: teams set the rules for agent-built work before it starts, and people judge what falls outside them.