Automate AMI Sorting and Waitlist Management for Lease-Up

Standard property management software like RealPage and AppFolio include application portals, but their native income calculation features are not built for the unique compliance requirements of affordable housing. These platforms typically use trailing 12-month data, which directly conflicts with the anticipated 12-month income projection required for LIHTC, HOME, and HUD programs. They struggle to reliably parse variable income sources such as hourly wages, tips, commissions, or bonuses, forcing leasing teams into extensive manual review processes for almost every applicant. This manual burden often creates significant bottlenecks during critical lease-up periods when hundreds or thousands of applications arrive in a short timeframe.

To compensate, leasing teams often resort to complex spreadsheets as a workaround. Consider a 500-unit lease-up expecting 1,500+ applications in the first few weeks. An agent must manually download pay stubs and other income documents from RealPage or AppFolio, project anticipated annual income (e.g., hourly wages multiplied by 2080 hours, plus estimated tips and bonuses), perform asset verification for HOME-layered units, check student status, and then manually assign each applicant to the correct AMI tier (30%, 40%, 50%, 60%, 70%, 80%) on a separate master spreadsheet. At an estimated 15-20 minutes per application, this workflow quickly accumulates into hundreds of hours of backlog, delaying the ability to even begin sorting the waitlist effectively.

This manual process is highly prone to errors. A simple typo in a spreadsheet formula can misclassify an applicant, leading to compliance violations, wasted time processing unqualified files, or inadvertently denying eligible applicants. Furthermore, RealPage and AppFolio cannot natively sort waitlists by income or AMI tier, meaning leasing staff must manually tag or cross-reference names, consuming 40+ hours per week for a team of 7+ during peak periods. This entire workflow is slow, does not scale to the demands of lease-up volume, and increases the overall denial rate due to processing delays or inaccuracies in pre-screening.

Syntora's engagement would commence with a detailed discovery phase to meticulously define your specific application workflow, identify all required integration points with your existing systems (such as RealPage or AppFolio APIs), and comprehensively document all applicable LIHTC, HOME, and HUD compliance rules, including specific AMI tables for your properties. This initial step ensures the engineered system aligns precisely with your operational needs and regulatory obligations.

Technically, the solution would involve building a custom integration layer designed to ingest new applicant data and documents in near real-time using your existing system's APIs. For robust document processing, we utilize large language models like the Claude API. We have extensive experience building document processing pipelines using Claude API for complex financial documents in other sectors, and the same robust pattern applies to efficiently extract key income figures—such as hourly wage, pay periods, tips, commissions, and bonuses—from various applicant documents, returning structured JSON data. This approach supports comprehensive income anticipation for the next 12 months, rather than relying on trailing income history.

A custom Python service, built with FastAPI, would house the core income calculation logic. This service would apply your specific LIHTC, HOME, or HUD rules to the extracted data, accurately projecting anticipated annual income. The calculated income would then be compared against your property's AMI tables, securely stored in a database like Supabase Postgres, to assign the correct AMI bucket and determine projected qualification status. This FastAPI service would be architected for event-driven processing and deployed on AWS Lambda, offering exceptional scalability for high-volume lease-up periods and optimizing hosting costs.

Upon a new application submission, a webhook would trigger the Lambda function to perform the calculation and parsing. The system would then use the RealPage API or AppFolio API to write the determined AMI tier, student status, asset verification triggers for HOME units, and pre-qualification status back to the applicant's record within your native system. This allows your leasing team to efficiently access auto-sorted waitlists by AMI tier directly within their standard software, eliminating manual tagging and spreadsheet work. For transparency and auditability, we integrate structured logging using tools like `structlog`, providing a complete, immutable trail of every calculation and decision. The system would also be designed to flag applicants for manual review if document parsing is ambiguous or income projections are uncertain, sending targeted notifications to designated channels with direct links to the relevant records, thereby focusing manual efforts where they are most needed and reducing denial rates caused by pre-screening errors.

A typical engineering engagement for a system of this complexity ranges from 12 to 20 weeks, with the timeline varying based on integration complexity and the specific nuances of your compliance rule variations. Key client contributions include providing secure API access to existing systems, detailed AMI tables, and comprehensive documentation of all applicable compliance rules. Deliverables would encompass the fully deployed custom automation system, complete technical documentation, and thorough training and handover for your operational and technical teams.