Anthropic engineers describe Claude Platform, Opus 5.5, and agent isolation
Original title: How Anthropic ships the #1 LLM in the World safely: Separating the Brain from the Hands
The episode covers managed-agent sandboxes, credential controls, API scaling, and safety routing rather than a new model drop.
Katelyn Lesse and Angela Jiang tell Latent Space how Anthropic ships the Claude Platform, including the Opus 5.5 launch, pricing, and evaluation. They describe separating the model’s “brain” from the agent’s “hands,” plus harnesses, sandboxes, credentials, and network controls for managed agents. A new intro covers Anthropic’s S-1, improvement and investment loops, and why the platform matters. Show notes point to launch notes and related write-ups; the item is an interview, not a standalone release.
- Guests discuss launching Opus 5.5, model behavior, evaluation, and pricing.
- Managed agents decouple the brain, harness, and sandbox.
- Isolation covers credentials, network controls, and long-running agent harnesses.
- Other topics include Messages API work in Rust and safety classifiers.
- A new introduction covers Anthropic’s S-1 and investment loops.
- Vendors
- Anthropic
- Organizations
- Latent Space
- AI models
- Claude Opus 5.5
From the original article
1 min read at latent.space
Katelyn Lesse and Angela Jiang join Latent Space to explain the engineering behind the Claude Platform: launching Opus 5. 5, separating the model’s “brain” from the agent’s “hands,” scaling APIs, and building reliable managed agents. A new introduction ...
By Latent.Space · latent.space
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.