Anthropic engineer says 70% of staff use self-improving loops to build agent graphs
blaizedsouza · x · 2026-07-27
An Anthropic engineer says 70% of the company’s engineers are using self-improving loops and predicts that agentic graphs will become the default way to orchestrate agents within 3–6 months.
The post points to a 30-minute lecture and article that walk through:
- building a self-improving agentic workflow from scratch
- why multi-step agents should be structured as graphs instead of a simple linear chain
- how to think about the role of a “graph architect” in agent engineering
Related event: Anthropic Engineers Detail Self-Improving Agent Loops(5 posts)→
More from coding & agent
- NVIDIA says Nemotron 3 Ultra leads open models in agentic RTL design — imjustnewatai · 2026-07-27
- A new @asciidotdev agent set up a sandbox, synced files and found the available CLIs — jasonkneen · 2026-07-27
- Claude Code’s ultracode mode powers a browser game build, but burns $2,500 in tokens — majidmanzarpour · 2026-07-27
- LangChain ships Deep Agents v0.7.0b2 with 50% fewer base tokens — LangChain · 2026-07-27
- Hermes Field Kit adds an open-source skill for interviewing users before starting a task — alexcovo_eth · 2026-07-27
- METR’s new metric measures when AI agents become more expensive than humans — The Decoder · 2026-07-27