Hackernews posts about Evals

1. FrontierCode: An eval to measure whether you would actually merge the code (cognition.ai)
   256 points by streamer45 13 days ago | 57 comments
2. The Vanta AI Quality Eval Maturity Model (www.vanta.com)
   26 points by hamelj 12 days ago | 2 comments
3. Show HN: Promptloop – create, run, and improve prompt evals from the terminal (github.com)
   13 points by velapod 24 days ago | 3 comments
4. Show HN: JavaScript Crossword – a crossword where the clue = eval(answer) (lyra.horse)
   7 points by rebane2001 28 days ago | discuss
5. Proof of AGI is the impossibility of evals (thewatershed.markpesce.com)
   4 points by mpesce about 23 hours ago | 2 comments
6. Pelican, or pelican't? A hint at Claude evals (noperator.dev)
   4 points by noperator 9 days ago | discuss
7. Zork-bench: An LLM reasoning eval based on text adventure games (www.lowimpactfruit.com)
   4 points by evakhoury 20 days ago | discuss
8. ProofLayer Rules – runtime security, red-team evals for LangGraph (github.com)
   3 points by dchitimalla1 9 days ago | 2 comments
9. SchemaFlow: Agentic Database Change Impact Analysis, SQL Gen and Eval Guardrails (developers.openai.com)
   3 points by gmays 8 days ago | discuss
10. Agent Judge: Solving Long-Context Evals for Production Agents (www.judgmentlabs.ai)
    3 points by gmays 11 days ago | discuss
11. Exercises in benchmarking, evals, and experimental design, part 6 (www.patreon.com)
    3 points by mfrw 14 days ago | discuss
12. Realistic Evals, or You're Blind (zozo123.github.io)
    3 points by zozo123-IL 18 days ago | discuss
13. JS Crossword: a crossword where the clue = eval(answer) (lyra.horse)
    3 points by thunderbong 28 days ago | discuss
14. What 50k Runs of a 5-Line Eval Taught Us (code.visualstudio.com)
    2 points by dahjelle 3 days ago | discuss
15. Command injection in NLTK collocations via eval() (aydinnyunus.github.io)
    2 points by yakkomajuri 12 days ago | discuss
16. Macro Evals for Agentic Systems (developers.openai.com)
    2 points by gmays 12 days ago | discuss
17. We had to build new evals for Fable (hex.tech)
    2 points by izzymiller 12 days ago | discuss
18. What happens when run you evals on brainrot? (www.scorecard.io)
    2 points by yash1hi 13 days ago | discuss
19. Reality: The Final Eval – Vending Bench Eval (www.latent.space)
    2 points by davedx 15 days ago | discuss
20. Show HN: HermesBench – workflow reliability evals for personal AI agents (verkyyi.github.io)
    2 points by verkyyi26 22 days ago | discuss
21. Marketing skill for Claude with 26 evals – +20pp over baseline (github.com)
    2 points by healman 22 days ago | discuss
22. Agent evals should feel like real work (www.zohaib.cc)
    2 points by zed_labs_dev 28 days ago | discuss
23. Bateschess – Chess Analytics Feeding Stockfish Evals into LLM's (bateschess.com)
    2 points by wtbates99 29 days ago | discuss
24. Show HN: TruLayer – tracing, evals, and a control loop for production LLMs (trulayer.ai)
    2 points by trulayer 30 days ago | discuss
25. Open-source AI skills that make Claude/ChatGPT produce real work, eval-scored (github.com)
    1 points by mohitagw 3 days ago | discuss
26. Build Your Own Eval Harness from Scratch with Bun and Claude -p (alexop.dev)
    1 points by speckx 5 days ago | discuss
27. GEDD – Grounded Eval-Driven Development for AI Agents (github.com)
    1 points by balasvce2026 10 days ago | discuss
28. Show HN: VQAScore – open eval metric/reward model, now for text-to-video (github.com)
    1 points by linzhiqiu 13 days ago | discuss
29. How to Debug AI Agents with Traces and Evals (medium.com)
    1 points by sukhpinder0804 18 days ago | discuss
30. JavaScript Crossword – a crossword where the clue = eval(answer) (lyra.horse)
    1 points by birdculture 26 days ago | discuss