Kanverse.ai: AI Invoice and Insurance Automation

Kanverse AI offers an AI-powered intelligent document processing platform for enterprises, specializing in invoice automation, insurance claims, and extraction from structured and unstructured documents.

Overview

Kanverse AI provides a hyperautomation platform that digitizes document processing from ingestion through filing. The company's multi-stage AI engine combines computer vision, machine learning, natural language processing, fuzzy logic, and OCR to automate document workflows across accounts payable, insurance, and financial compliance use cases.

The company's growth strategy centers on embedding within established enterprise platforms rather than competing against them. The January 2023 Kunzite release extended the platform beyond AP automation into handwriting recognition, signature detection, and KYC/KYB processing. In April 2023, Kanverse joined the Boomi Technology Partner Program, with CRO Sameer Deshpande emphasizing zero-touch document processing for enterprise business processes. By December 2023, the company launched a validated Guidewire Marketplace integration for PolicyCenter users, positioning itself inside the insurance platform rather than alongside it.

That distribution logic carried into 2025 and 2026. Kanverse listed its AP invoice automation on Oracle Cloud Marketplace in July 2025, making it directly accessible to Oracle Fusion Cloud ERP, E-Business Suite, and NetSuite customers. The company also joined the Oracle Applications & Technology Users Group (OATUG), embedding within Oracle's user community. CEO Karan Yaramada stated the move was about "making it easier than ever for enterprises to harness the power of AI to eliminate manual processes, reduce costs, and scale with confidence." The Oracle listing is a distribution play; no new capabilities were announced alongside it, and pricing remains undisclosed. Competing IDP vendors with Oracle Marketplace presence, including ABBYY and Rossum, are not addressed in available coverage, leaving Kanverse's relative standing among Oracle-native AP automation options unestablished.

The next major product milestone is the Unakite release, scheduled for April 20, 2026, which introduces a conversational AP Assistant, embedded AI agents, and an LLM-based classification framework. This signals a shift from template-based extraction toward zero-shot, conversational interfaces, aligning Kanverse with broader industry movement toward agentic document processing.

How Kanverse.ai processes documents

Kanverse AI's multi-stage AI engine combines computer vision, machine learning, NLP, and fuzzy logic across a pipeline that runs from ingestion through validation and filing. The platform ingests documents from email, portals, and scanners, then applies pre-trained AI models to extract structured data before routing for approval or filing.

Handwriting recognition and signature detection, added in the Kunzite release, extend processing to complex documents that defeat traditional OCR. Adaptive capabilities automatically adjust to document format changes without manual retraining. Business metrics tracking provides visibility into paid and unpaid invoices, early payment discounts, and employee productivity.

The April 2026 Unakite release adds a conversational AP Assistant that lets finance teams query invoice data and generate custom reports through natural language, with alerts routed through MS Teams and Google Chat. A new Vendor Agent named "Veronica" integrates DocuSign for e-signatures and WhatsApp for vendor messaging. The release also introduces an LLM-based classification framework supporting zero-shot and few-shot text classification without extensive retraining, a meaningful shift from the rule-based and template-dependent approaches that characterized earlier versions. Platform enhancements include advanced filter functionality with custom date ranges, FTP/SharePoint export, stagnant document detection, and a redesigned Line Item Viewer with line-level numbering and multi-shipment PO mapping.

The platform claims up to 99.5% extraction accuracy and, in the context of the Oracle listing, up to 80% reduction in invoice processing cycle times. Both figures are vendor-stated. The Fellowes Brands deployment, where the platform processes over 100,000 invoices annually across 17 international subsidiaries without template configuration, provides the closest available third-party validation of the accuracy claim, though Fellowes has not published independent benchmark data.

Use cases

AP invoice automation

Organizations deploy Kanverse to automate invoice processing from receipt through payment approval. The platform extracts line items, vendor information, and financial data from diverse invoice formats, including Excel, PDF, Word, and image files, validating against purchase orders and business rules before routing for approval.

The Fellowes Brands case study is the most concrete public evidence of this capability at scale. Fellowes processes approximately 100,000 invoices annually across North American AP operations. Prior template-based systems delivered "very low extraction accuracy," according to the case study. Kanverse resolved the accuracy problem without requiring template configuration, demonstrating template-free extraction across 17 international subsidiaries where invoice formats vary by supplier and region.

The Oracle Cloud Marketplace listing extends this capability to Oracle Fusion Cloud ERP, E-Business Suite, NetSuite, and hybrid ERP environments, covering PO matching, compliance validations, and audit trails. Kanverse claims up to 80% reduction in invoice processing cycle times for Oracle customers; this figure is vendor-stated and has not been corroborated by a named reference customer outside of Kanverse's own materials.

For a broader view of AP automation approaches and benchmarks, see the accounts payable automation guide.

Insurance claims processing

Insurance carriers use Kanverse to process ACORD forms, custom forms, loss run reports, statement of values, and quotation request slips. CPO Dr. Akhil Sahai described the problem directly: "The inconvenient, highly manual submission intake process is tedious, time consuming, error-prone, and causes unprecedented delays and bottlenecks for insurers." The Guidewire-validated integration for PolicyCenter places Kanverse inside the insurer's existing workflow rather than requiring a separate integration project. Guidewire's Will Murphy, VP of Global Technology Alliances, noted that "Kanverse modernizes document intake for insurers" and that making AI the cornerstone of its technology "ensures accuracy and adaptability as the industry evolves."

Kanverse claims 99.5% accuracy on insurance document extraction and reduction of submission intake cycle times from days or weeks to minutes. Both figures come from Guidewire and Kanverse press materials; no independent insurer has published corroborating data.

For context on how IDP fits into broader claims workflows, see the insurance claims processing guide.

KYC/KYB document processing

Financial institutions implement Kanverse for customer onboarding workflows. The Kunzite release added pre-trained AI models accessible through APIs for processing identity documents, financial statements, and compliance paperwork. The handwriting recognition capability added in the same release is particularly relevant for onboarding documents that include handwritten fields, a common gap in template-based OCR systems.

Technical specifications

Resources

• Website
• Documentation
• Fellowes Brands case study
• Oracle Cloud Marketplace listing announcement
• Unakite release notes

Company information

Headquarters: San Jose, California, United States

Product releases: Unakite (April 2026), Kunzite (2023), Iolite (2022), Fluorite (2021)

Key executives: Karan Yaramada (CEO), Sameer Deshpande (CRO), Dr. Akhil Sahai (CPO)

Partner ecosystem: Oracle (Cloud Marketplace + OATUG membership), Guidewire (validated marketplace accelerator), Boomi (Technology Partner Program), UiPath, DocuSign