Anthropic Bets on Agent Ecosystem Over Walled Garden
Training Data (Sequoia) · rss · 2026-07-14
Anthropic's Katelyn Lesse and Angela Jiang discuss building an ecosystem rather than a "walled garden," where external developers and Anthropic's own products build on the same developer platform.
They abstract the platform into three layers: knowledge, execution, and coordination. The real leverage lies in the top "strategies" layer—a meta-orchestration mechanism where tokens take on different roles like suggesting, executing, reflecting, or remembering. They also mention partnering with Modal, Vercel, and Cloudflare for self-hosted sandboxes, emphasizing the importance of standardized connectors and MCP specs.
A key stance is their willingness to open up capabilities like skills and MCP to the entire industry, allowing Claude and non-Claude agents to interconnect. However, model routing remains closed, as they believe harnesses are best tuned for a single model family rather than generalizing across models.
More from coding & agent
- Alex Townsend posts 200 open problems in numerical linear algebra for humans and AI agents — IgorCarron · 2026-09-11
- Kimi K2.8 Preview rolls out: near-K3 coding performance, 1M context for all tiers — teortaxesTex · 2026-09-11
- Looking for a classifier of software engineering task shapes to pick models per task — StewartalsopIII · 2026-09-11
- Steal this idea: prompt-to-hardware where agents assemble custom devices — paraschopra · 2026-09-11
- Model Is the Least Interesting Part: A Guide to Six Core AI Architectures from RAG to Multi-Agent — goyalshaliniuk · 2026-09-11
- Non-coder builds layered memory architecture: 20k tokens tracks a year of agent conversations — matteoianni · 2026-09-11