Two humans, two Claude Code agents: 'my Claude tells his Claude what to fix'
Hefty_Professor_4170 · reddit · 2026-10-22
A developer who built a coding harness for a company describes 30+ Git repos accessible to a fleet of Claude Code agents that pick up tagged Jira tickets, clarify requirements with reporters, and implement. Then came the absurd moment: his Claude Code reviewed a QA colleague's tickets and gave feedback — while the colleague used his own Claude Code to respond and update them. Two humans ended up as message relays between AI agents, before yet another set of agents picked up the tickets to write code. "I can't decide whether this is a glimpse into the future of software engineering or a completely ridiculous way to get things done. Maybe both."
More from coding & agent
- Solo dev splits MacBook, Mac mini and a server to run his posting agents — petrusenko_max · 2026-10-22
- Dev builds LLM-chat trading for Seer prediction markets, open-sources the bot — devanshmehta · 2026-10-22
- Anthropic safety guardrails block LLM-driven crypto trading agent, dev reports — devanshmehta · 2026-10-22
- "Understanding debt": vibe coding fails the moment anything breaks — gerardsans · 2026-10-22
- mdancho84: The Valuable Ones Architect Systems — Agents, Workflows, Memory, Eval Loops — mdancho84 · 2026-10-22
- AutoResearch: AI Agents That Run the Full Research Loop Autonomously — mdancho84 · 2026-10-22