https://www.youngju.dev/blog 천천히 올바르게. AI Researcher & DevOps Engineer Youngju's tech blog. GPU/CUDA, LLM, MLOps, Kubernetes AI workloads, distributed training, and data engineering. ko fjvbn2003@gmail.com (Youngju Kim) fjvbn2003@gmail.com (Youngju Kim) Fri, 12 Jun 2026 00:00:00 GMT https://www.youngju.dev/blog/ai/2026-06-12-loop-engineering-agentic-coding-systems.en https://www.youngju.dev/blog/ai/2026-06-12-loop-engineering-agentic-coding-systems.en The era of manually prompting an agent every turn is ending. This post covers plan-act-verify-self-correct loop design, verification loops that wire tests and linters into agent feedback, how to design repo-level guide files like CLAUDE.md, multi-agent division of labor, checkpoints for long autonomous runs, and failure modes such as loop runaway and reward hacking — all from a hands-on perspective. Fri, 12 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) ailoop-engineeringagentsclaude-codedevopsautomation https://www.youngju.dev/blog/ai/2026-06-12-loop-engineering-agentic-coding-systems.ja https://www.youngju.dev/blog/ai/2026-06-12-loop-engineering-agentic-coding-systems.ja エージェントに毎ターン手動でプロンプトを入力する時代は終わりつつあります。plan-act-verify-self-correctのループ設計、テストとリンターをエージェントのフィードバックにつなぐ検証ループ、CLAUDE.mdのようなリポジトリレベルのガイドファイル設計法、マルチエージェント分業、長時間自律作業のチェックポイント、ループ暴走や報酬ハッキングといった失敗モードまで、実戦の観点から扱います。 Fri, 12 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) ailoop-engineeringagentsclaude-codedevopsautomation https://www.youngju.dev/blog/ai/2026-06-12-loop-engineering-agentic-coding-systems https://www.youngju.dev/blog/ai/2026-06-12-loop-engineering-agentic-coding-systems 에이전트에게 매 턴 수동으로 프롬프트를 입력하는 시대가 끝나가고 있습니다. plan-act-verify-self-correct 루프 설계, 테스트와 린트를 에이전트 피드백으로 연결하는 검증 루프, CLAUDE.md 같은 리포 수준 가이드 파일 설계법, 멀티 에이전트 분업, 장시간 자율 작업의 체크포인트, 루프 폭주와 보상 해킹 같은 실패 모드까지 실전 관점에서 다룹니다. Fri, 12 Jun 2026 00:00:00 GMT fjvbn2003@gmail.com (Youngju Kim) ailoop-engineeringagentsclaude-codedevopsautomation