piyushptiwari/insureos-training-data

INSUREOS Training Data — Insurance AI Dataset

Created by [Bytical AI](https://bytical.ai) — AI agents that run insurance operations.

Dataset Description

A comprehensive synthetic training dataset for UK insurance AI models. Covers six task types across the complete insurance operations lifecycle.

Files

SFT Data Schema

{
  "instruction": "What is the FCA Consumer Duty and how does it affect UK insurers?",
  "response": "The FCA Consumer Duty (PS22/9) came into force on 31 July 2023...",
  "category": "regulation"
}

DPO Data Schema

{
  "prompt": "Explain the role of Lloyd's syndicates in the London market.",
  "chosen": "Lloyd's syndicates are the risk-bearing entities within Lloyd's of London...",
  "rejected": "Lloyd's is a big insurance company in London..."
}

Document Classification Labels (12 classes)

Policy Schedule, Certificate of Insurance, Claim Form, Loss Adjuster Report, Bordereaux — Premium, Bordereaux — Claims, Endorsement, Renewal Notice, Statement of Fact, FNOL Report, Subrogation Notice, Policy Wording

NER Entity Types (13)

CLAIMNUMBER, DATE, INSURER, LOB, MGA, MONEY, ORG, PERIL, PERSON, POLICYNUMBER, POSTCODE, REGULATION, SYNDICATE, VEHICLE

Coverage Topics

• —Regulation: FCA Consumer Duty, GDPR, Solvency II, IDD, Senior Managers Regime
• —Market: Lloyd's of London, London Market, MGA operations, reinsurance
• —Claims: FNOL, triage, reserving, subrogation, fraud indicators
• —Underwriting: Risk assessment, pricing factors, exposure management
• —Operations: Bordereaux processing, premium collection, policy administration

How to Use

from datasets import load_dataset

# Load SFT data
sft = load_dataset("piyushptiwari/insureos-training-data", data_files="insurance_sft_10k.jsonl")

# Load claims data
claims = load_dataset("piyushptiwari/insureos-training-data", data_files="claims_motor_25000k.csv")

Learn How These Models Work

Want to understand how this data trains real ML models? We wrote a plain-English guide that walks through every INSUREOS model — the concepts, the data, and the training code:

• —Learning guide: LEARN.md
• —Training & data-generation code: piyushptiwari/insureos-models — see the `data/` generators and the `training/` scripts

Part of the INSUREOS Model Suite

Built by Bytical AI for training the INSUREOS model family:

• —InsureLLM-4B — Insurance domain LLM
• —InsureDocClassifier — Document classification
• —InsureNER — Named Entity Recognition
• —InsureFraudNet — Fraud detection
• —InsurePricing — Premium pricing

About Bytical AI

Bytical builds AI agents that run insurance operations — claims automation, underwriting intelligence, digital sales, and core system modernization for insurers across the UK and Europe. Microsoft AI Partner | NVIDIA | Salesforce.