← Back to Dashboard
Pipeline Leaderboard
🎮 Blind Taste TestPlay
| # | Pipeline | Blind Win Rate | Avg Score | Liability | Credibility | Damages | ★ Rating | Runs |
|---|---|---|---|---|---|---|---|---|
| 🏆 | Testing new Pinecone Indexing | 100% | 4.09 | 4.00 | 3.87 | 4.39 | — | 1 |
| #2 | Production Pipeline | 100% | 3.94 | 4.21 | 3.66 | 3.94 | — | 2 |
| #3 | Pinecone Vector Indexing | 57% | 4.08 | 4.09 | 3.86 | 4.28 | — | 5 |
| #4 | Testing Claude | 57% | 4.06 | 4.16 | 3.85 | 4.15 | ★★★★(4.0) | 12 |
| #5 | Legal Document Pipeline | 50% | 4.02 | 4.05 | 3.89 | 4.12 | ★★★★(4.0) | 11 |
| #6 | Legal Document Pipeline copy | 20% | 3.99 | 4.00 | 3.79 | 4.19 | ★★★(3.0) | 5 |
| #7 | NEW - Legal Document Pipeline | 0% | 4.09 | 4.15 | 3.67 | 4.45 | — | 2 |
| #8 | Test Using Codex as Agent | 0% | 4.02 | 4.09 | 3.77 | 4.19 | — | 1 |
| #9 | Offer Studio | No games yet | 3.85 | 3.84 | 3.52 | 4.20 | — | 69 |
Blind win rate = head-to-head wins ÷ total comparisons across all Blind Taste Test games. Pipelines with no game history are ranked by avg computed score.