ingest.py

from app.services.document_loader import load_pdf
from app.services.embedding_service import get_embedding_model
from app.services.vector_store import create_vector_store

# =========================================================
# LOAD DOCUMENT
# =========================================================

# SPLIT DOCUMENT INTO CHUNKS
chunks = load_pdf(
    "app/data/company_docs/ZOperationPolicies.pdf"
)

# =========================================================
# LOAD EMBEDDING MODEL
# =========================================================

embedding_model = get_embedding_model()

# =========================================================
# CREATE VECTOR DATABASE
# =========================================================

create_vector_store(chunks, embedding_model)

print("Knowledge base created successfully.")