This PR publishes two new essays for the BeyondTheCode blog, grounded in signals from Hacker News.

• The selected HN post links:

• https://martinfowler.com/fragments/2026-02-18.html
• https://cepr.org/voxeu/columns/how-ai-affecting-productivity-and-jobs-europe
• https://georgeguimaraes.com/your-agent-orchestrator-is-just-a-bad-clone-of-elixir/
• https://resobscura.substack.com/p/what-is-happening-to-writing

• Why each topic cleared the selection bar:

• "The Supervisory Engineering Middle Loop" was selected because it identifies a critical shift in the Individual Contributor role—from implementation to supervision—that is often invisible to organizational measurement systems. It addresses a durable structural shift in how impact is defined under AI acceleration.
• "The Architectural Inevitability of Agents" was selected because it surfaces the recurring confusion between traditional web development patterns and the stateful, non-deterministic requirements of AI agents, grounding modern challenges in 40-year-old architectural solutions.

• The conceptual gap identified:

• The divergence between visible velocity (Output Capital) and the invisible rigor of constraint management (Judgment Capital).
• The "maturity gap" where modern developers reinvent distributed systems primitives in runtimes not designed for them.

• Which essay corresponds to which post:

• "The Supervisory Engineering Middle Loop" draws signal from Post 14 and Post 18.
• "The Architectural Inevitability of Agents" draws signal from Post 30.

• 2–4 short quotations (max 25 words each) from different commenters:

• "Code is now like steel. It's somewhat valuable by itself, but we don't need the town blacksmith to make us things anymore."
• "100% of the time spent in agent frameworks is spent ... waiting for the agent to respond, or waiting for a tool call to execute."
• "AIslop is everywhere, and seemingly unavoidable for companies to use more and more to cut more corners."
• "I don't really remember Claude 3.5 doing this, but it seems increasingly worse, with 4.6 being so bad I don't like using it for brainstorming."

PR created automatically by Jules for task 15304161364478346753 started by @rockoder