Skip to content

Conversation

@Lotte-Verheyden
Copy link
Contributor

@Lotte-Verheyden Lotte-Verheyden commented Nov 18, 2025

Important

Adds a script for running QA chatbot experiments, refactors chatbot handling, and updates dependencies.

  • New Script:
    • Adds run-chatbot-experiment.ts to execute QA chatbot experiments with Langfuse and OpenAI.
    • Configures experiments with datasets general-questions and irrelevant-questions.
    • Includes evaluators like isConcise, whyExplanation, isUnderstandable, etc.
  • Refactoring:
    • Moves prompt processing logic to chatbotCore.ts with getProcessedPrompt() and chatbotStep().
    • Updates apiHandler.ts to use chatbotStep() for handling chatbot requests.
  • Dependencies:
    • Adds dotenv and tsx to package.json.
    • Updates @langfuse packages to version 4.4.2.
    • Adds experiment:chatbot script to package.json.

This description was created by Ellipsis for 3df3092. You can customize this summary. It will automatically update as commits are pushed.

- Moved core chatbot logic out of handler function, into separate functions, into their own file
Add script to run chatbot experiments on datasets with AI evaluators:
- Created run-chatbot-experiment.ts which runs experiments on multiple datasets (general-questions, irrelevant-questions)
- Moved flush logic from chatbotCore to apiHandler for better separation
@vercel
Copy link

vercel bot commented Nov 18, 2025

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Preview Updated (UTC)
langfuse-docs Canceled Canceled Nov 18, 2025 5:41pm

@dosubot dosubot bot added size:XL This PR changes 500-999 lines, ignoring generated files. enhancement New feature or request labels Nov 18, 2025
Lotte-Verheyden and others added 2 commits November 18, 2025 18:21
Co-authored-by: ellipsis-dev[bot] <65095814+ellipsis-dev[bot]@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

enhancement New feature or request size:XL This PR changes 500-999 lines, ignoring generated files.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants