Hello from Greg, Max and the Parsewise team! 👋

TL;DR

Parsewise turns multiple documents into a single structured response. Cross-document entity linking, contradiction detection, bounding boxes and UI toolkit to show it all included.

See it live!

The Problem

Today, you have to build and maintain complex document processing pipelines with changing business rules. You parse, classify, and rely on structured responses from LLMs or IDP tools (e.g., Reducto) to get individual extraction results that you piece together with other bits of information. There’s no reliable way to catch when information contradicts itself, which is risky. Finally, you build a custom verification UI for your operations team to deal with LLM mistakes.

The Solution

We provide an API to abstract all that away into a single call. You provide multiple documents and the desired output to get back a response with resolved values, flagged contradictions, and full traceability across documents / pages that you can display in your own app.

Get Started

=======================================================

Sign up with free credits: https://www.parsewise.ai/get-started
  (use the Agents.md for a 1m integration)

=======================================================

Story Time

Why are we working on this, how we got to where we are, who we are working with, and where we are headed.

Greg was at Palantir trying to make sense of various Japanese insurance documents and realizing that any processing pipeline built around it was highly custom, time-consuming, and brittle. Max was on the human operator side at Bain & Co, realizing that the technology was being held back due to a trust problem.

We realized that there is this powerful new capability for general reasoning (and thus processing). Models are getting extremely good at interpreting complex documents (shout-out to the Gemini model family) and perform better than most OCR-type solutions on the market.

So we set out to build the layer that comes on top:
1. Provide mapping from user intent to a dynamic ontology
2. Use the now-abundant intelligence to reason across documents to get to the right results
3. Provide the provenance trace and validation UX that is required in many use cases

We are working with leading customers, both small and large, like UBS, Compre Group, Thinksurance, and more. They rely on Parsewise to streamline submission intake, claim portfolio risk assessment, etc., for both API based integrations and operational workflows.

With this launch, we are inviting developers to build more reliable and delightful workflows around their unstructured data processing. As we scale, we are excited to be helping with the end-to-end integration and deterministically filling forms, mapping to 3rd party APIs and web portals.

See more at https://www.parsewise.ai/api