Claude Code’s Creator Stopped Prompting Claude — He Writes Loops and Merges 150 PRs a Day From His Phone
Last Updated on June 14, 2026 by Editorial Team
Author(s): Chew Loong Nian – AI ENGINEER
Originally published on Towards AI.
Claude Code's Creator Stopped Prompting Claude — He Writes Loops and Merges 150 PRs a Day From His Phone
Boris Cherny, the person who created Claude Code, hasn’t hand-written a line of code since October 2025. On one record day he merged 150 pull requests. He runs five to ten Claude sessions at a time, spawns a few hundred agents across them during the day, and lets a few thousand run overnight on deeper work. His primary development machine is his phone.
After the lead, the article explains the shift from “prompting” to “writing loops”: instead of crafting better instructions each time, you design a recurring process that schedules Claude, checks outcomes (e.g., whether CI is red or green), and reruns automatically until it succeeds—moving the human from the inner loop to the outer one. It then details what Cherny’s setup looks like in practice: a phone-first workflow with multiple concurrent sessions, sub-agents for delegated work, and continuous loop jobs for tasks like PR handling and CI repair, while noting that his scale and access at Anthropic make direct replication unrealistic. The piece breaks down the available scheduling primitives (session-bound loops vs. time schedules vs. cloud/off-machine tasks vs. headless cron) and offers a concrete five-minute pattern to build a basic “babysit CI” cron loop using headless mode and safeguards like max-turns. Finally, it weighs caveats—skill and correctness still matter, costs can be high, and pure throughput doesn’t guarantee maintainable merges—and concludes that the main takeaway is to start small with one verifiable loop task you can trust.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Related posts
Recent Posts
