Document integrations for Moroku Flow
AI-Powered OCR and LLMs in Loan Origination
The landscape of loan origination is undergoing rapid transformation, driven by advances in the market and technology, Traditional lending workflows, often reliant on manual document review and rigid rule-based systems, are being replaced by intelligent automation that can ingest, interpret, and validate unstructured data from PDFs, scanned forms, and email attachments with unprecedented accuracy.
Recent developments in Generative AI, Vision Transformers, and multimodal LLMs have enabled systems to not only extract text but also understand document context, layout, and semantics. Models such as Donut, LayoutLMv3, and TrOCR are leading the charge, offering capabilities like:
- End-to-end document parsing without explicit OCR pre-processing
- Context-aware field extraction from diverse layouts
- Semantic validation and anomaly detection for fraud prevention
These models are increasingly being embedded into Intelligent Document Processing (IDP) platforms such as Affinda, Docsumo, and Klippa, which expose their capabilities via APIs and SDKs for seamless integration.
Broker & Market Lead Channels: Unlocking Flexibility Through AI-Powered Ingestion
- 75% of all new residential loans in Australia were arranged by brokers in 2024, up from 57% in 2017
- This figure is expected to reach 80% by the end of 2025, according to Loan Market CEO David McQueen
- The broker channel contributes $4.1 billion in economic activity and supports over 37,000 jobs
Key Implications:
- Rapid onboarding of new channels: Brokers and aggregators can be activated without custom API builds.
- Dynamic form mapping: AI models like Donut and LayoutLMv3 interpret layout and semantics, enabling field-level extraction from unfamiliar formats.
- Reduced integration overhead: No need for bilateral API contracts or middleware for each partner.
- Improved data quality: AI validation and enrichment ensures completeness and consistency before submission to credit decisioning engines.
- On-demand scalability: New lead sources can be trialled and scaled without engineering bottlenecks.
Pluggable Integration: Moroku Lending’s Strategic Advantage
- Rapid onboarding of third-party AI services for OCR, fraud detection, and document classification
- Toggle-based activation of specific providers or models depending on document type, geography, or compliance needs
- Low-code orchestration of workflows across Vue.js and Node.js components, allowing dynamic routing of documents to the most appropriate AI engine
- Scalable experimentation with emerging LLMs and OCR tools without vendor lock-in or replatforming.
Approaches - Library or API?
OCR & AI API Libraries: The Developer's Toolkit
Tesseract, TrOCR, Donut, and LayoutLMv3, offering deep control and customisation. They’re typically hosted within the Moroku infrastructure and used to build bespoke document processing workflows. Characteristics:
- Code-level control over extraction logic, model tuning, and deployment
- Ideal for embedding into Node.js or Python microservices within Moroku’s orchestration layer
- Require in-house resources for configuration, scaling, and compliance
- Suitable for edge cases like proprietary form formats or offline processing
Examples:
| Library | Description | Integration Notes |
|---|---|---|
| Tesseract OCR | Mature open-source OCR engine maintained by Google | Best for clean printed text; can be wrapped in Node.js or Python microservices |
| TrOCR (Microsoft) | Transformer-based OCR model for high-accuracy text recognition | Available via Hugging Face; ideal for structured document parsing |
| LayoutLMv2 / LayoutXLM | Document understanding models that combine OCR output with layout and semantics | Requires OCR pre-processing; excellent for form field extraction |
| Donut (NAVER) | End-to-end OCR-free document parser using Vision Transformers | Outputs structured data directly (e.g. JSON); ideal for loan forms and contracts |
Web Services & OCR APIs: The Plug-and-Play Platforms
Affinda, Klippa, Docsumo, and Artificio, offering hosted endpoints that ingest PDFs, scanned files, or emails and return parsed, structured data. Characteristics:
- Quick to integrate via RESTful APIs, with minimal setup or training overhead
- Often bundled with extra features like fraud detection, validation, or mobile OCR SDKs
- Compliance and scalability handled by provider (e.g. GDPR, ISO 27001)
- Offer pay-per-use models or tiered subscriptions for cost predictability
Examples:
| Service | Key Features | Integration Potential |
|---|---|---|
| Affinda | AI OCR for loan applications, supports PDF/email ingestion, 20+ fields extracted | REST API, bulk upload, supports 56+ languages |
| Klippa DocHorizon | OCR + data extraction for financial documents, including loan forms | Offers SDKs, JSON/XML/CSV output, mobile scanning |
| Artificio | End-to-end loan processing automation with AI OCR, NER, and validation | Email inbox integration, custom ML models, ERP connectors |
| Docsumo | Intelligent document processing for loan forms, bank statements, and ID docs | Real-time extraction, fraud detection, credit scoring support |
| Algodocs | IDP platform for loan document parsing and structured data output | OCR + NLP + ML stack; supports scanned and digital formats |
Key Differences at a Glance
| Feature | API Libraries | Web Services & OCR APIs |
|---|---|---|
| Control & Customisation | High | Moderate to Low |
| Setup Time | Longer | Minimal |
| Scalability & Hosting | Self-managed | Vendor-managed |
| Compliance Burden | Internal | Outsourced |
| Flexibility for Unique Forms | High | Varies by provider |
| Cost Model | Free/Open Source + infra cost | SaaS-style licensing |
Conclusion: The Shifting Landscape of Loan Origination
Key Market Shifts
- Broker dominance: Brokers now originate ~75–80% of residential loans in Australia, with aggregators like LMG, AFG, and Finsure leading the charge.
- Marketplace origination: Platforms like Lendi, Uno, and property portals (e.g. Domain, REA Group) embed lending flows into real estate journeys.
- Fintech disruption: Startups leverage AI, blockchain, and real-time data to streamline application and approval processes.
- PropTech lending: Real estate-focused fintechs offer alternative credit models, fractional ownership, and embedded finance.
Implications for Borrowers
- Greater channel diversity: Borrowers can initiate applications from brokers, marketplaces, or embedded flows, each with unique formats and data structures.
- Fragmented documentation: Application forms vary widely across channels, often lacking standardisation or API compatibility.
- Increased data portability: Open Banking and AI-powered OCR enable borrowers to share financial data securely and dynamically.
Integration & Document Consumption Strategies
- Pluggable ingestion layers: Loan origination platforms like Moroku can ingest PDFs, emails, and scanned forms using AI-powered OCR and LLMs, bypassing rigid API mappings.
- Channel-agnostic orchestration: Modular systems allow dynamic routing of documents to appropriate AI engines for parsing and validation.
- Rapid partner onboarding: Brokers and marketplaces can be integrated without custom builds, accelerating time-to-market and reducing friction.
- Semantic normalisation: LLMs like Donut and LayoutLMv3 interpret layout and context, enabling field-level extraction from diverse formats.
Scale across fragmented origination channels
| Aggregator | Brokers | Loan Book | Notes |
|---|---|---|---|
| Loan Market Group - LMG | 6,000+ | $370B+ | Largest Aggregator across AU/NZ |
| Australian Finance Group FMG | 3,000+ | $160B+ | Strong tech stack and panel lender |
| Finsure | 2,500+ | $85B+ | Known for CRM and rapid growth |
| Connective | 3,500+ | $100B+ | Independent Model with Mercury Nexus platform |
| Mortgage Choice | 1,000+ | $80B+ | Owned by REA Group, strong brand presence |