A curated collection of practical AI projects implementing OCR systems, RAG, AI agents, and other AI use cases.
"""
RAG Agent with Database Routing - Qdrant database setup.
Three in-memory Qdrant collections are created (empty) on startup.
Documents are added by the user through the UI.
"""
from __future__ import annotations
import uuid
from qdrant_client import QdrantClient
from qdrant_client.models import Distance, PointStruct, VectorParams
from .embeddings import OrqEmbeddings
VECTOR_SIZE = 3072 # google-ai/gemini-embedding-001 output dimension via Orq.ai
COLLECTIONS = ["products", "support", "financial"]
SCORE_THRESHOLD = 0.5
def build_databases(orq_api_key: str) -> tuple[QdrantClient, OrqEmbeddings]:
"""Create empty in-memory Qdrant collections."""
client = QdrantClient(":memory:")
embeddings = OrqEmbeddings(orq_api_key)
for collection_name in COLLECTIONS:
client.create_collection(
collection_name=collection_name,
vectors_config=VectorParams(size=VECTOR_SIZE, distance=Distance.COSINE),
)
return client, embeddings
def add_documents(
client: QdrantClient,
embeddings: OrqEmbeddings,
collection_name: str,
texts: list[str],
) -> int:
"""Embed and insert a list of text chunks into the given collection.
Returns the number of documents successfully added.
"""
texts = [t.strip() for t in texts if t.strip()]
if not texts:
return 0
vectors = embeddings.embed_documents(texts)
points = [
PointStruct(
id=str(uuid.uuid4()),
vector=vector,
payload={"text": text, "source": collection_name},
)
for text, vector in zip(texts, vectors)
]
client.upsert(collection_name=collection_name, points=points)
return len(points)
def doc_count(client: QdrantClient, collection_name: str) -> int:
"""Return the number of vectors stored in a collection."""
info = client.get_collection(collection_name)
return info.points_count or 0