The AI coding gap: Limestone ships 98% AI-written code while enterprises lag
alex_verem · x · 2026-11-18
About 98% of Limestone's code isn't handwritten, while many enterprise teams are barely past autocomplete. The author argues the gap isn't the tool — everyone has the same models — and frames AI coding maturity in four levels: (1) AI suggests lines, engineer writes code; (2) AI writes full PRs, humans review each; (3) AI also reviews, humans decide merges; (4) some changes merge without human eyes. Most mid-market teams are at level 1; the author runs at 2–3 and won't rush to level 4 on systems that bill patients or move money. Trust comes from engineering practices like tests (details truncated in the repost).
More from coding & agent
- Antipode: An Open-Source Local Synth Controlled via MCP, Turning Agent Ideas into WAVs — Dizzy_Basil6892 · 2026-11-18
- Founder builds real-time voice support agent with Claude and Agora's AI Developer Toolkit — heyshrutimishra · 2026-11-18
- Audio EQ and compression inside Antigravity: Gemini 4 builds in-chat FFmpeg console — alexanderchen · 2026-11-18
- Auto-approving PRs: Qodo's rules-based workflow handles the coding-agent PR bottleneck — omarsar0 · 2026-11-18
- Dev influencer slams "AI dreaming" claims: it's just a random semantic search DB — gerardsans · 2026-11-18
- Record your judgment process 3 times, let AI build business evals from it — alliekmiller · 2026-11-18