Commit Graph

2 Commits

Author SHA1 Message Date
eligrinfeld
ce97671da3 test: add CI/CD workflow 2025-01-05 14:16:31 -07:00
eligrinfeld
6bcee39e63 feat(cleanup): Enhanced business data validation and cleaning
- Added confidence scoring system for data quality
- Implemented strict validation for emails, phones, and addresses
- Added batch processing to prevent LLM overload
- Improved error handling and fallback mechanisms
- Added caching based on confidence scores

Technical changes:
- Added regex validation for contact info
- Implemented scoring system (0-1 scale)
- Added timeout protection for LLM calls
- Enhanced post-processing for consistent formatting
- Added business type detection for context

Breaking changes: None
Dependencies: No new dependencies required
2025-01-04 20:59:00 -07:00