Document Intelligence Search

from mixpeek import Mixpeek

client = Mixpeek(api_key="YOUR_API_KEY")

namespace = client.namespaces.create(name="doc-search")
collection = client.collections.create(
    namespace_id=namespace.id,
    name="contracts",
    extractors=["pdf-extraction", "text-embedding-v2", "ocr"],
    chunk_strategy="page-based"
)

# Upload documents
client.buckets.upload(
    collection_id=collection.id,
    url="s3://your-bucket/contracts/"
)

# Search with high BM25 weight for exact legal terms
results = client.retrievers.execute(
    retriever_id=retriever.id,
    query="indemnification clause with liability cap"
)