Vibe coding: how LanChess wrote 100,000 lines in 3 months
LanChess is a rare, honest case study of what happens after vibe coding. The service's author built a production chess analytics product in under three months: 3,300 prompts, 832 commits and 100,000 lines of AI-written code. But the main takeaway is not about speed. It is that what comes next is personal data, authorization, infrastructure and ongoing support.
AI-processed from Habr AI; edited by Hamidun News
LanChess has become a clear case study in vibe coding: how an AI prototype turns into a real production service with users, a backend and legal constraints.
How do you build a production service with AI in 3 months?
The LanChess case shows that using an AI assistant for vibe coding (3,300 prompts, 832 commits) produced 100,000 lines of code. But the key point is that real engineering was still needed (backend, legal frameworks), along with solutions to problems that a single request to a model will not solve.
How did
LanChess manage to write 100,000 lines of code in 3 months? The author used vibe coding with 3,300 prompts, which resulted in 832 commits and let him quickly turn a prototype into a full production service.
- Nov 1, 2026Local models lose to Opus and Codex: 0.21 on a MacBook versus 1.00 for Codex
- Nov 1, 2026Kimi scores 96.2% on a 1C database analytics test and emerges as a fallback if Claude access is blocked
- Nov 1, 2026OpenAI Rewrites GPT Behavior Rules: Five Values and a Chain of Command
- Oct 31, 2026Author puts his script before an AI jury: the verdict cost 850 thousand tokens
Key Questions Answered
How do you build a production service with AI in 3 months?
The LanChess case shows that using an AI assistant for vibe coding (3,300 prompts, 832 commits) produced 100,000 lines of code. But the key point is that real engineering was still needed (backend, legal frameworks), along with solutions to problems that a single request to a model will not solve.
How did LanChess manage to write 100,000 lines of code in 3 months?
The author used vibe coding with 3,300 prompts, which resulted in 832 commits and let him quickly turn a prototype into a full production service.
Test Frontier LLMs & Autonomous Agents in One Workspace
Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.