Automate Affordable Housing Application Review

Most teams start with generic form builders like Jotform. They collect applications, but the real work of downloading PDFs, calculating income, and checking AMI limits remains manual. It creates a digital pile of paperwork, not a workflow. The leasing team is still buried in administrative tasks that prevent them from filling units.

Property Management Systems like RealPage and AppFolio have workflow tools, but they are built for market-rate housing. They handle basic income verification but fail on the specific LIHTC requirement for anticipating the next 12 months of income from hourly or inconsistent sources. They cannot automatically parse a photo of a pay stub to extract hours, rate, and YTD figures. This forces teams into a "swivel chair" workflow, copying data from the PMS into a separate compliance spreadsheet.

A leasing team for a 500-unit LIHTC lease-up receives 2,000 applications in the first week. Using their PMS, an agent must open each application, find the pay stubs, use a calculator to annualize income (e.g., $18/hr x 40 hours x 52 weeks), then check that total against a printed AMI chart. This takes 10-15 minutes per applicant, creating a 333-hour backlog just to build the initial waitlists.

Syntora's approach would begin with a discovery phase to understand your current application intake and qualification workflow. We would identify the specific APIs for your RealPage or AppFolio instance to ingest new applications, or define an alternative secure intake method for documents.

For income documents, the system would use the Claude API to parse PDFs and JPGs of pay stubs, offer letters, and benefit statements. This technology extracts line items such as hourly rate, hours per week, and YTD earnings, returning structured JSON data. This process is designed to eliminate manual data entry.

The structured data would feed into a Python service, typically running on AWS Lambda for scalability and cost efficiency. We would implement custom functions to codify your specific LIHTC and HOME rules for anticipating income, calculating factors like annualized hourly wages, averaged regular tips, and seasonal bonuses. The system would then calculate the total anticipated 12-month income, compare it against the property's specific AMI table loaded from a Supabase database, and assign the correct AMI bucket.

Once an applicant is sorted, the system would update their status in AppFolio or RealPage via API, placing them onto the correct AMI-tiered waitlist. We can also integrate an automated email trigger to acknowledge receipt and provide a projected qualification status, aiming to reduce follow-up inquiries. The goal is for your leasing team to work from a pre-sorted list of qualified applicants.

The backend of such a system is typically built with FastAPI, deployed on AWS Lambda. The Supabase database would hold configuration data such as AMI tables and property-specific rules, designed to be updatable without requiring code changes. As a key deliverable of the engagement, Syntora provides a custom-built system, fully documented in a runbook, and stored in your private code repository, granting you full ownership and control.