{"slug": "i-built-an-ai-judge-that-doesn-t-decide-who-wins", "title": "I Built an AI Judge That Doesn't Decide Who Wins.", "summary": "A developer built FairJudge AI, an AI-assisted hackathon evaluation system that scores projects against a rubric using evidence-based evaluation, deterministic Python scoring, and multi-judge comparison with disagreement detection. The system deliberately stops short of naming a winner, leaving the final decision to human judges, and was created to address inconsistent rubric application across judges. The project was submitted to the Hacktoberfest Weekend Challenge: Build for a Friend.", "body_md": "*This is a submission for the [Hacktoberfest Weekend Challenge: Build for a Friend](https://dev.to/challenges/hacktoberfest-weekend-2026-10-01)*\n\nMy friend participates in hackathons frequently and builds strong projects, but she had an important concern:\n\n**How can you know that judges are applying the same rubric consistently?**\n\nA project can receive different scores from different judges even when everyone is evaluating the same criteria.\n\nSo I built **FairJudge AI**.\n\nFairJudge is an AI-assisted hackathon evaluation system designed to make judging more **transparent, evidence-based, and consistent**.\n\nIt can:\n\nBut there is one important thing FairJudge **doesn't** do:\n\n**It does not decide who the objectively correct winner is.**\n\nThe final decision always remains with human judges.\n\nHackathon judging involves multiple judges evaluating projects across several criteria.\n\nDisagreements can happen because:\n\nInstead of building another AI that simply says:\n\n\"Project X should win.\"\n\nI wanted to build something more transparent:\n\n**Why was this score given? What evidence supports it? What evidence is missing? Where do judges disagree?**\n\nThat became **FairJudge AI**.\n\n🎥 **Demo Video:**\n\n[https://claude.ai/artifact/9JLZQKa7C33ADuJg2Qk5wC](https://claude.ai/artifact/9JLZQKa7C33ADuJg2Qk5wC)\n\nThe demo shows:\n\nThe core workflow is:\n\n```\ntext\nProject + Rubric\n       ↓\nEvidence-based AI Evaluation\n       ↓\nEvidence Validation\n       ↓\nDeterministic Python Scoring\n       ↓\nMultiple Judge Comparison\n       ↓\nDisagreement Detection\n       ↓\nHuman Review\n```\n\n", "url": "https://wpnews.pro/news/i-built-an-ai-judge-that-doesn-t-decide-who-wins", "canonical_source": "https://dev.to/dikshit_garg/i-built-an-ai-judge-that-doesnt-decide-who-wins-58h9", "published_at": "2026-10-04 18:34:39+00:00", "updated_at": "2026-10-04 18:43:05.374032+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-tools", "ai-agents"], "entities": ["FairJudge AI", "Hacktoberfest"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/i-built-an-ai-judge-that-doesn-t-decide-who-wins", "markdown": "https://wpnews.pro/news/i-built-an-ai-judge-that-doesn-t-decide-who-wins.md", "text": "https://wpnews.pro/news/i-built-an-ai-judge-that-doesn-t-decide-who-wins.txt", "jsonld": "https://wpnews.pro/news/i-built-an-ai-judge-that-doesn-t-decide-who-wins.jsonld"}}