FD-300-proto: Build synthetic user generator for testing #10
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This framework enables automated testing of chatbot conversations using LLM-generated synthetic users with specific behavioral traits. Key components: - personas.py: 5 user personas (evasive, oversharer, date_confused, emotionally_flooded, matter_of_fact) with detailed system prompts - simulator.py: Conversation loop that alternates between real chatbot and synthetic user responses - evaluators.py: Quality checks for robotic patterns (banned phrases, echoing, repetition, question balance, variety) - conftest.py: pytest fixtures and markers (--synthetic, --synthetic-llm) - test_quality_evaluators.py: 19 unit tests for evaluator logic - test_synthetic_conversations.py: Integration tests with mocked and real LLM modes Usage: - Unit tests (no LLM): pytest btcopilot/tests/synthetic/test_quality_evaluators.py - Mocked tests: pytest --synthetic btcopilot/tests/synthetic/ - Full LLM tests: pytest --synthetic --synthetic-llm btcopilot/tests/synthetic/
Adds a framework to evaluate whether conversations collect enough family data as defined in the system prompts. New files: - btcopilot/personal/data_requirements.py: Single source of truth for data collection requirements (presenting problem, family of origin, extended family, own family, nodal events). Includes MINIMUM_COMPLETE_REQUIREMENTS defining when data collection is "done". - btcopilot/tests/synthetic/data_completeness.py: Evaluator that checks conversations against requirements using heuristics or LLM analysis. - btcopilot/tests/synthetic/test_data_completeness.py: 20 tests for requirements validation and completeness evaluation. The data_requirements module can generate checklist markdown matching the prompts, ensuring the prompt checklist and test criteria stay in sync. Usage: - Unit tests (heuristics): pytest btcopilot/tests/synthetic/test_data_completeness.py - LLM tests: pytest --synthetic --synthetic-llm btcopilot/tests/synthetic/
patrickkidd
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.