idcohen/tearsheets

Build mock tearsheets of banking clients from made up data

Tearsheets

Build mock tearsheets of banking clients from made up data

First pass (6/22): Answer simple client questions given various resources.

• Build mock data (linkedin, bank facts, transactions) for one client
• Using an LLM, determine which data categories are relevant to an input question
• Produce the answer by embedding question and relevant data into a second prompt

6/23 meeting recap: The below WatsonX.ai generative platform resources will provide you with hands-on experience of the platform's capabilities:

• Watsonx.ai Workbench: Access the workbench to try out the promptlab.
• Watsonx GitHub: Explore the GitHub repository to gain how to use the ibm-generative-ai SDK
• Pitchbook Description Extraction: For insights into Pitchbook description extraction, refer to the Pitchbook-Description notebook.