{"slug": "agent-engineering-claude-code", "title": "Agent engineering: Claude Code", "summary": "A developer describes shifting from using LLMs as occasional third-party dependencies to adopting an agent-driven execution model where the agent loop serves as the primary environment, relegating deterministic code like Python to a minor role for structured storage and API calls.", "body_md": "I started with LLMs as another third-party dependency. In my apps, code defined the logic, and LLM calls handled small things like summarizing or tagging input. Over time, as agents gained more of my trust, I started experimenting with turning the execution model inside out.\n\nInstead of code that occasionally calls an LLM, I use the agent loop as the execution environment itself. The agent drives the flow, and deterministic code only shows up where I actually need hard logic: structured storage and external API calls. Instead of being the main driver, Python becomes a small fraction of the overall logic.", "url": "https://wpnews.pro/news/agent-engineering-claude-code", "canonical_source": "https://roman.pt/posts/claude-code-as-your-execution-environment/", "published_at": "2026-06-02 00:00:00+00:00", "updated_at": "2026-06-15 20:42:00.821227+00:00", "lang": "en", "topics": ["large-language-models", "ai-agents", "ai-tools"], "entities": ["Claude Code"], "alternates": {"html": "https://wpnews.pro/news/agent-engineering-claude-code", "markdown": "https://wpnews.pro/news/agent-engineering-claude-code.md", "text": "https://wpnews.pro/news/agent-engineering-claude-code.txt", "jsonld": "https://wpnews.pro/news/agent-engineering-claude-code.jsonld"}}