How cloudera/dataserviceslabs works
This repository is a collection of Cloudera Data Services workshops, not one deployed application. In the sampled AgenticAI workflow, a customer submits a request through Gradio; a CrewAI manager routes it to policy, promotion, or order-feedback agents, which use their tools to retrieve knowledge or query customer data. A separate chatbot workshop retrieves context from a vector index and uses an LLM to answer questions. The tree also includes data-engineering lab scripts. README capabilities for CDW and CDF lack corresponding runtime files in the supplied tree, so they are not drawn. Arrows reflect sampled code or explicit documentation; other workshop wiring is not inferred.
Main entry points: agents_and_tasks.py Routing manager · vector_rag_utils.py Policy retrieval · llm_app_final.py Knowledge chatbot · graph_query_wrapper.py Customer graph access
Components 22
Agentic customer service
- Support UI
ClouderaAI/AgenticAI/chat_app/agents_app_launch.py - Crew orchestrator
ClouderaAI/AgenticAI/chat_app/agents_app_launch.py - Routing manager
ClouderaAI/AgenticAI/chat_app/crew_agents/agents_and_tasks.py - Policy agent
ClouderaAI/AgenticAI/chat_app/crew_agents/agents_and_tasks.py - Sales agent
ClouderaAI/AgenticAI/chat_app/crew_agents/agents_and_tasks.py - Feedback agent
ClouderaAI/AgenticAI/chat_app/crew_agents/agents_and_tasks.py - Policy lookup
ClouderaAI/AgenticAI/chat_app/tools/retrieve_policy_tool.py - Promotion lookup
ClouderaAI/AgenticAI/chat_app/tools/targeted_promos_tool.py - Feedback submission
ClouderaAI/AgenticAI/chat_app/tools/feedback_submission_tool.py - Customer graph access
ClouderaAI/AgenticAI/chat_app/chat_utils/graph_query_wrapper.py
Knowledge chatbots
- Policy retrieval
ClouderaAI/AgenticAI/chat_app/chat_utils/vector_rag_utils.py - Vector index ingestion
ClouderaAI/LLMChatbot-GenAI/LLM-hol-user-guide/2_populate_vector_db/pinecone_vectordb_insert.py - Knowledge chatbot
ClouderaAI/LLMChatbot-GenAI/LLM-hol-user-guide/8_launch_app_final/llm_app_final.py
Data engineering labs
- Data extraction and reporting
ClouderaDataEngineering/Resources/Scripts/Lab3B1_Data_Extraction_Sub_150k.py - Large-data extraction
ClouderaDataEngineering/Resources/Scripts/Lab3B2_Data_Extraction_Over_150k.py - Report creation
ClouderaDataEngineering/Resources/Scripts/Lab3B3_Create_Reports.py - Airflow lab
ClouderaDataEngineering/Resources/Scripts/Lab5_airflow_dag.py
Other components
- Customer
- Customer graph
- Vector index
- Knowledge files
- Language model service
Connections 20
- Customer → Support UI: submits request
- Support UI → Crew orchestrator: launches crew
- Crew orchestrator → Routing manager: routes request
- Routing manager → Policy agent: delegates
- Routing manager → Sales agent: delegates
- Routing manager → Feedback agent: delegates
- Policy agent → Policy lookup: looks up policy
- Policy lookup → Policy retrieval: searches context
- Policy retrieval → Vector index: queries vectors
- Policy retrieval → Knowledge files: reads matched file
- Sales agent → Promotion lookup: gets promotions
- Promotion lookup → Customer graph access: queries customer
- Feedback agent → Feedback submission: submits feedback
- Feedback submission → Customer graph access: writes feedback
- Customer graph access → Customer graph: runs graph queries
- Vector index ingestion → Knowledge files: reads documents
- Vector index ingestion → Vector index: upserts embeddings
- Customer → Knowledge chatbot: asks question
- Knowledge chatbot → Vector index: retrieves context
- Knowledge chatbot → Language model service: invokes model