Checking a Florida contractor properly meant searching state records, confirming insurance and reading reviews, so most homeowners skipped it. We built My Contractor Report: it pulls licensing, business filings, workers' comp coverage, BBB status and reviews into one AI risk-scored PDF report.
Amplence builds AI-powered data platforms that turn scattered, multi-source lookups into clear, actionable reports.
Homeowners and property managers in Florida had no simple way to check a contractor before hiring. Doing it properly meant searching the state licensing database, looking up the company's corporate filings, checking workers' compensation coverage, reading BBB and customer reviews, and working out which permits the job needed. It was slow, the records were hard to read, and most people skipped it. Our client needed a platform that could gather these records automatically, apply clear risk scoring, and hand back a professional report anyone could read and act on.
For each order, the platform pulls the contractor's Florida DBPR license record, SunBiz corporate filing, workers' compensation coverage and exemptions, BBB profile and public reviews. Each source runs as its own tracked check, so customers can follow the status of every one, and orders that need a closer look are moved to manual review.

Raw records mean little to a homeowner. An AI layer reads the collected data, scores each area (license, business status, reviews, BBB profile, coverage and exemptions) and gives an overall risk level in plain language. It also maps the customer's project scope to the permits and Notice of Commencement the job is likely to need.
Every completed order becomes a branded PDF report of 20+ pages: an executive summary with sources, licensing and business records, review sentiment, permit guidance and the risk assessment. Customers download it from their dashboard and can raise a dispute if something looks wrong.
Customers order reports and track each data check. Contractors get a paid portal with QR tools, report visibility controls and activity tracking. Admins manage the order queue, disputes, reviews, contractor accounts and analytics.

The platform is built with React and Supabase, which handles authentication, user data, report storage and dashboard state, with Stripe handling payments.
Each data source runs as its own tracked check with its own status, so every order shows exactly which checks are complete and which need attention. Once the checks finish, the results are combined into one record, the AI layer scores it, and the PDF report is generated.
The report lists the sources behind its findings, so readers can check them. If a customer thinks something is wrong, they can open a dispute, and disputes are reviewed within 1 to 2 business days.

Each source is collected by its own pipeline: Florida DBPR licensing, SunBiz corporate filings, workers' compensation records, BBB and public reviews. The results are cleaned and combined into one record, and each source is tracked as a separate check on the order, so a problem is visible instead of silently missing.

The AI layer scores each area separately (license, business status, customer reviews, BBB profile, coverage and exemptions) and combines them into an overall risk level with a plain-language summary. The report shows where each piece of data came from, so the reader can check it.

Each report is a dated snapshot of the records at the time it was run. When a contractor's situation changes, a new report pulls the current records. Customers can also dispute anything that looks wrong, and disputes are reviewed within 1 to 2 business days.

Yes. The admin portal shows the order queue with the status of every data check, plus disputes, reviews, contractor accounts and analytics, so the team can run the platform day to day without engineering help.

This platform took 3 months from scoping to launch, built on React and Supabase. Most of the engineering time went into the data work: collecting and reconciling records that were never designed to be read by software, so the AI layer works from clean, consistent data.