Skip to content

Dataset · snorkelai

Multi-turn insurance underwriter AI assistant dialogues

snorkelai/Multi-Turn-Insurance-Underwriting

Examples of ReAct agent operations in small business commercial insurance tasks by Snorkel AI.

What's inside

A small dataset (under 1000 entries) in English combining tabular data and text. It contains traces of multi-step dialogues between an insurance specialist and an AI assistant built using LangGraph, MCP, and ReAct agents. Each example reflects solving a specific task for evaluating small business insurance applications.

What you can do with it

  • Design and debug ReAct agents for financial and insurance analytics
  • Evaluate the quality of multi-step AI reasoning when working with tabular applications
Tasks
Question answering
Languages
English
Size
up to 1K rows
Downloads this month
1.2K
Likes
37

How to load it

from datasets import load_dataset

ds = load_dataset("snorkelai/Multi-Turn-Insurance-Underwriting")