How Anthropic uses Claude Code: Agentic Software Engineering at Scale - Daisy Hollman

Daisy Hollman

Talks worth your time · NDC Conferences · 2026-09-29 · 60 min · 11 pages

This talk was recorded at NDC Copenhagen in Copenhagen, Denmark. #ndccopenhagen #ndcconferences #developer #softwaredeveloper Attend the next NDC conference near you: https://ndcconferences.com https://ndccopenhagen.com/ Subscribe to our YouTube channel and learn every day: / @NDC Follow our Social Media! https://www.facebook.com/ndcconferences https://twitter.com/NDC_Conferences https://www.instagram.com/ndc_conferences/ #ai #claudecode Over the past year, the capabilities of coding agents have expanded massively, and software engineers are demanding more of them than ever. But context window sizes have pretty much stagnated—one million tokens was roughly the state of the art a year ago, and it remains so today. As we move towards a world where agents regularly tackle monorepo-scale tasks, the demand to squeeze more and more information into those fixed limits of model context has grown into a real, full-time engineering task. In this talk, I'll walk through the design of Claude Code plugins and their constituent context engineering primitives. For each one, we'll ask the crucial question for software engineering in 2026: does it scale? We'll talk about how we're answering this question for our own code at Anthropic—both in terms of what's working and what isn't.

Generated automatically from the recording's transcript. It's meant to replace re-watching, not the recording itself — verify against the video at the timestamp before quoting formally. Slides belong to the speakers and organisers. Produced with recording-to-pdf.