In May 2026, Anthropic held their Code with Claude developer conference in London. From the main stage, an Anthropic engineer stated: "Most software at Anthropic is now written by Claude. Claude has written most of the code in Claude Code." Spotify presented what this looks like operationally — not as an experiment, but as standard practice. Their top engineers stopped writing code manually in December 2025. Delivery Hero's autonomous AI agent, Herogen, now merges over 100 pull requests every day — the equivalent annual output of a typical engineer, produced daily. These are not pilots. They are not proof-of-concept projects running in a separate team. They are the org chart. And most succession plans were built for the org chart that existed before them.

Here is the problem the calibration room has not named yet.

The AI restructuring happening in software is not a technology story. It is a leadership readiness story in disguise — and most succession plans are not equipped to see it.

When Spotify integrated Claude Code into their engineering infrastructure, they did not just change a tool. They changed the role. Engineers who once wrote code now manage agents that write code. The skills that made someone succession-ready in 2024 — deep technical execution, structured problem-solving within defined systems — are now table stakes at best. The role has been redesigned around a different capability profile.

Delivery Hero went further. Their autonomous agent Herogen now produces the daily coding output of 130 senior engineers. When one in five merged pull requests is agent-generated, the human engineers remaining are not doing less of the same job. They are doing a fundamentally different one.

I have watched this dynamic play out across functions for thirty years. Technology changes the job before the calibration room catches up. The result is a succession pipeline full of people who are ready for a role that no longer exists — and a gap at the front of the bench for the role that does.

The AI-readiness question for a CHRO is not "are your leaders using AI tools?" Most of them are. The real question is whether your succession nominees are ready to lead in an org that has been redesigned around AI as the primary capability. That is a different question. Most calibration sessions are still answering the first one.

Which of your succession nominees were calibrated against role criteria that predate the AI restructuring of their function — and when were those criteria last reviewed?

In your next calibration session, how will you distinguish between a leader who is ready for the role as it exists today, and one who is ready for the role as it will exist in 18 months?

How Spotify cut migration time by 90% with the Claude Agent SDK — The official account of Spotify's integration: 650+ agent-generated pull requests merged per month, 90% reduction in engineering time on complex code migrations. Read it as a talent story, not a technology one — what changed is not the tooling, it is the job. → https://claude.com/customers/spotify

Delivery Hero unveils Herogen — autonomous AI agent unlocks 130-person engineering output — Delivery Hero's public announcement: 100+ merged PRs daily, 85% success rate, targeting deployment to 50% of 4,000+ engineers by year-end 2026. The org chart implication is buried in the numbers. (Read the VP of Platform quote: "Agentic engineering replaces rigid programming syntax with the fluidity of natural language.") → https://www.deliveryhero.com/newsroom/delivery-hero-unveils-herogen-autonomous-ai-agent-unlocks-130-person-engineering-output/

MIT Technology Review — Anthropic's Code with Claude showed off coding's future, whether you like it or not — The best single account of the London event: what was said on stage, who was there, and what the shift from "AI pilot" to "AI org chart" actually looks like in practice. The context for why the succession conversation needs updating. → https://www.technologyreview.com/2026/05/21/1137735/anthropics-code-with-claude-showed-off-codings-future-whether-you-like-it-or-not/

Edition 3 — 77% of CEOs say their exec team isn't AI-ready. What does that mean for your succession bench? — Edition 3 framed the warning signs. Edition 7 is about what happens when the restructuring is already complete. Worth reading together. (TTI edition) → https://newsletter.totalia.ai/p/77-of-ceos-say-their-exec-team-isn-t-ai-ready-what-does-that-mean-for-your-succession-bench-total-ta

The AI restructuring is moving through functions in sequence. Software engineering is the leading indicator. The CHRO who updates their succession criteria now — before the restructuring reaches their function — is the one whose board conversation looks different.

They can subscribe here: https://newsletter.totalia.ai