The context
Quoting vehicle insurance in Colombia usually involves several systems that don't talk to each other: the advisor looks up the plate in RUNT, cross-references Fasecolda data, then manually re-enters much of that same information into each insurer's own system to get a rate. It's a process repeated many times a day, with high risk of data-entry error, and it depends on the advisor's patience to jump between platforms.
The solution
Polis automates that entire flow in four steps: the advisor enters the vehicle's plate number; the system resolves it against RUNT and pulls make, model, year and technical data automatically; the client is asked only for the minimum required information (insured party, date of birth, city, contact details); and Polis chains the insurer's internal calls (rate, package validation, capacity) to generate an official quote, ready to download as a PDF.
This is a case of AI process automation applied to a highly regulated, system-fragmented industry — not computer vision like Spray Analyzer, but API orchestration and the elimination of repetitive manual work across platforms that were never designed to talk to each other.
The result
Figures published by the product itself at poliscolombia.com, not our own estimates.
Being honest about where the project stands
Unlike Spray Analyzer, which already runs with a client in consolidated production, Polis is in active beta (v0.1) with one insurer integrated live. We include it as a case precisely because it's real and working — not a mockup — but it's honest to say it's still growing: more insurers and more volume are the next step, not an established fact yet.
Technology used
Integration and orchestration of external APIs (RUNT, Fasecolda, insurer systems), a backend that chains calls and validates data at each step, and automatic generation of official quote PDFs — the same robust backend and automation approach we apply to any project that needs to connect multiple external systems.
Why we include it
Spray Analyzer demonstrates our computer vision capability. Polis demonstrates the other half: integrating external systems and automating complex administrative processes, in a regulated industry, with real usage data — exactly the kind of project we describe in our AI process automation service.