Insurance quote generation and policy comparison
A broker-facing platform that parses policy documents with an LLM, builds explainable premium quotes, versions them, and compares quotes side by side.
The challenge
Insurance agents rekey customer and coverage details from PDF, Word and JSON policies into quoting tools, then explain pricing differences to customers. Without versioning and an audit trail, revisions to quotes are hard to trace.
What we built
Agents upload policy documents which are parsed by format-specific parsers, with an LLM on AWS Bedrock extracting structured customer and coverage data. A premium calculator applies coverage, deductible and risk multipliers and returns a breakdown of how the figure was reached. Quotes are stored with version history and can be restored, saved as templates, and compared side by side with filtering, sorting and export. The LLM also drafts quote summaries and customer emails, and answers questions about a quote. Authentication uses bcrypt hashing, sessions and account lockout after repeated failures.
Key capabilities
- Policy ingestion from PDF, DOCX and JSON with LLM-based field extraction
- Risk-based premium calculation with itemised breakdown
- Quote version history with restore and reusable templates
- Multi-quote comparison with attribute filtering, sorting and export
- LLM-generated quote summaries, customer emails and quote Q&A
- Session-based authentication with failed-login lockout
What it delivers
- Premiums are returned with a breakdown, so agents can explain how a figure was built
- Quote changes are versioned and restorable, giving a traceable revision history
- Core services (auth, parsing, premium calculation, comparison, export) have dedicated unit tests
- Schema changes are managed through numbered Alembic migrations
Related work
Insurance claims assessment with deterministic gates
An agentic pipeline that classifies motor and medical-aid claim PDFs, reconciles every figure in plain code, and returns a structured review plus an annotated PDF.
Read the case study →
Healthcare claims audit, denial prediction and appeals
A multi-tenant SaaS prototype that audits medical coding, predicts claim denials and drafts appeal letters, with payer rules and EHR (FHIR) connectivity.
Read the case study →
Invoice extraction with review and ERP sync
A vision-model invoice extraction tool with side-by-side human review, duplicate detection, mailbox ingestion and supplier-validated sync to accounting systems.
Read the case study →Have a similar backlog?
Most of our work started as a single painful workflow. Tell us yours.