configs/compute_rag_vector_index_meetings.yaml

# Meeting Data RAG Configuration for Online App
# This file is for using meeting data from test_meetings collection

parameters:
  extract_collection_name: test_meetings  # Source: our BigQuery meeting data
  fetch_limit: 100  # Process all meeting documents (we have ~100)
  load_collection_name: rag  # Destination: contextualized chunks
  content_quality_score_threshold: 0.5  # Lower threshold for meetings
  retriever_type: parent  # Use parent retriever (performed better in our tests)
  embedding_model_id: text-embedding-3-small
  embedding_model_type: openai
  embedding_model_dim: 1536
  chunk_size: 640
  mock: false
  processing_batch_size: 8
  processing_max_workers: 4
  device: mps  # Use MPS for Apple Silicon acceleration