Checking your AI assistant's quality across several measures

The scenario runs your AI assistant over a set of examples and scores its answers across several quality measures.

  • AI quality across several measures at once
  • Correctness, helpfulness, category and more
  • Scores in numbers
  • A summary can be emailed

How it works

To connect Brevo and , you don't need a developer: a ready-made scenario links them in minutes.

  1. Starts when: When fetching a dataset row
  2. Then: World Series Searcher
  3. Then: Evaluation3
  4. Then: Evaluation
  5. Then: Evaluation5
  6. Then: Google Gemini Chat Model
  7. Starts when: When fetching a dataset row1
  8. Then: Evaluation4
  9. Then: Evaluation6
  10. Then: Evaluation7
  11. Then: Evaluation8
  12. Then: Evaluation1
  13. Then: Evaluation10
  14. Then: Evaluation9
  15. Starts when: When fetching a dataset row2
  16. Then: World Series Searcher1
  17. Then: Evaluation2
  18. Then: Evaluation12
  19. Then: Evaluation11
  20. Starts when: When fetching a dataset row3
  21. Then: US Postal Formatter
  22. Then: Evaluation13
  23. Then: Evaluation15
  24. Then: Evaluation14
  25. Starts when: When fetching a dataset row4
  26. Then: Article Reviewer
  27. Then: Google Gemini Chat Model3
  28. Then: Evaluation16
  29. Then: Evaluation18
  30. Then: Evaluation17
  31. Starts when: When fetching a dataset row5
  32. Then: Product Description Rewriter
  33. Then: Google Gemini Chat Model4
  34. Then: Anthropic Chat Model
  35. Then: Perplexity Tool
  36. Then: Anthropic Chat Model1
  37. Then: OpenAI Chat Model
  38. Then: OpenAI Chat Model1
  39. Starts when: When fetching a dataset row6
  40. Then: Information Extractor
  41. Then: Evaluation20
  42. Then: Evaluation19
  43. Then: Evaluation21
  44. Check: If
  45. If yes: Send a message
  46. If no: Send a message1
  47. Starts when: Gmail Trigger
  48. Then: Extract from File
  49. Then: Google Gemini Chat Model1

You can launch this integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.