On the success of 'natural language programming' - Marc's Blog

kept by eddie dec 17, 25

Natural language programming succeeds because software has always been built through conversational specification loops, and LLMs now include computers in those loops.

Coding Agents Meet Distributed Reality
AI-generated distributed code should target frameworks like Hydro that make common concurrency bugs compile-time errors rather than runtime failures.
also on TLA+, #programming-languages, #ai-coding, #software-engineering, #developer-tools
AI will make formal verification go mainstream
Formal verification will become mainstream as AI coding agents leverage proof systems like Lean for validation, according to a Hacker News discussion.
also on TLA+, #programming-languages, #ai-coding, #software-engineering, #developer-tools
One agent, every surface: how we built the Kiro agent harness - Kiro
Kiro unified its IDE, CLI, web, and iOS agent codebases into a single server‑side harness using ACP, enabling cross‑client sessions and shared features.
also on Kiro, #ai-coding, #software-engineering, #developer-tools
What you're actually writing when you write a SKILL.md
Skills are loader specifications with three execution stages, not static prompts, and understanding this runtime architecture is the key to writing efficient, portable agent skills.
also on Kiro, #ai-coding, #software-engineering, #developer-tools
10 anti-AI slop moves for frontend projects going faster than humans can review—Martian Chronicles, Evil Martians’ team blog
Ten defensive practices—contract generation, strict TypeScript, behavior‑filter linting, layer boundaries, custom linter rules, mutation testing, dead‑code detection, duplicate detection, and CI enforcement—make AI‑generated frontend code safer and cheaper to review.
also on OpenAPI, #programming-languages, #ai-coding, #developer-tools