Tenant Screening Platform

Part of the worlds-biggest-software-project initiative.

An AI-native, FCRA- and fair-housing-aware tenant screening platform that gives landlords and property managers explainable risk decisions without bureau-grade pricing or compliance risk.

The Tenant Screening Platform is an open alternative to closed commercial screening tools such as TransUnion SmartMove, CoreLogic SafeRent, and AppFolio FolioScreen. It combines credit, criminal, eviction, and income data into an explainable risk score, automates FCRA adverse action workflows, and detects document fraud — built for independent landlords, mid-market property managers, and proptech platforms that need to embed screening as a feature.

Why Tenant Screening Platform?

Incumbent per-report fees ($15–$55) and enterprise subscriptions ($280+/month for AppFolio) price small landlords out of bureau-grade screening while offering little workflow automation.
Credit-bureau products (SmartMove, RentBureau, SafeRent) deliver data but provide weak compliance tooling: adverse action letters, fair housing checks, and audit trails are largely manual.
Algorithmic bias is a documented risk — CoreLogic/SafeRent faced a class action over its scoring methodology, and HUD has confirmed Fair Housing Act disparate impact theory applies to AI screening tools.
Document fraud is a real and growing problem: Snappt's 2025 data across 5M+ analyses found 6.4% of rental applications contained manipulated documents, yet fraud detection is typically gated behind enterprise contracts.
Existing tools rarely expose model-level explainability; landlords cannot defend automated decisions in fair-housing litigation without an audit trail tying score factors to disclosed adverse action reasons.

Key Features

Screening Data and Reports

Credit report retrieval via bureau partner API (TransUnion or Experian)
National multi-jurisdictional criminal background check
Eviction history report with address cross-referencing
Secure report delivery and 5-year audit-trail storage aligned with FCRA retention expectations
Reusable / portable applicant report architecture aligned with emerging state mandates

AI Risk Scoring and Decisioning

Explainable risk score synthesising credit, eviction, rental history, and income inputs
Factor-level disclosure that maps directly into FCRA-compliant adverse action reasons
Plain-language screening criteria builder with automated fair-housing pre-check
Disparate impact simulation for proposed policy changes
Automated decisioning with exception routing for edge cases needing human review

Compliance Automation

FCRA-compliant written consent collection and storage
Automated pre-adverse and adverse action letter generation with the FCRA Summary of Rights attachment
State- and city-specific compliance rules engine (ban-the-box, source-of-income / Section 8, portable report acceptance)
HUD-aligned individualised assessment workflow for criminal records
Inline legal guidance during screening setup

Fraud Detection and Identity

AI document fraud detection on pay stubs and bank statements (image and metadata signals)
Synthetic identity and AI-generated document detection
Government ID verification with tamper and liveness checks
Open banking income verification (Plaid / Mastercard Open Finance) instead of submitted pay stubs

Workflow and Integration

Applicant invitation and online application flow
Portfolio-level configuration of screening criteria
Listing and PM-suite integration paths for embedding screening as a step
API for proptech platforms embedding screening as a feature

AI-Native Advantage

Unlike incumbents that bolt AI onto bureau data, this project treats AI as the integration layer. A single explainable risk model fuses credit, rental history, bank-transaction income, and behavioural signals; a natural-language policy engine lets landlords describe screening criteria in plain English and flags fair-housing risk before a policy is deployed; document-image and statistical-anomaly models catch altered pay stubs, fake IDs, and synthetic identities; and adverse action letters are auto-drafted from the same factors that drove the decision — closing the explainability gap that fuels fair-housing litigation against algorithmic screeners.

Tech Stack & Deployment

The platform is designed as an API-first service with a landlord/PM web UI on top, suitable for self-hosted deployment by property management companies and embedded use by proptech platforms. Core integrations target bureau APIs (TransUnion, Experian Connect / RentBureau), open banking providers (Plaid, Mastercard Open Finance / Finicity), and PM suites (AppFolio, Buildium, Yardi via partner channels). Where standards exist — MISMO Rental Data Standards for rental payment exchange, FCRA disclosure formats, and HUD individualised-assessment guidance — the platform aligns with them rather than inventing proprietary equivalents.

Market Context

The tenant screening services market was valued at approximately $1.85 billion in 2025 and is growing at roughly 10–12% CAGR, sitting inside a broader background screening market of about $14.7 billion (Business Research Insights 2025; Mordor Intelligence 2025). Per-report pricing among incumbents ranges from $15 to $55, with enterprise PM suites such as AppFolio starting at around $280/month. Primary buyers are independent landlords (1–10 units), mid-market property managers (50–500 units), large institutional operators (500+ units) needing decisioning engines and audit trails, and proptech platforms embedding screening as a feature.

Project Status

This project is in the research and specification phase.
Contributions, feedback, and domain expertise are welcome.

Contributing

We welcome contributions from developers, domain experts, and potential users. See CONTRIBUTING.md for guidelines.

Important: All contributions must be your own original work or clearly attributed open-source material with a compatible licence. Copyright infringement and licence violations will not be tolerated and will result in immediate removal of the offending contribution. If you are unsure whether a piece of code, text, or other material is safe to contribute, open an issue and ask before submitting.

Licence

Licence to be determined. See discussion for context.