Glib.ai — BFSI Document Processing and Fraud Detection

6 min read

On This Page

Glib.ai is an AI-powered document processing platform built exclusively for banking, financial services, and insurance (BFSI) workflows, with integrated fraud detection and compliance automation. In early 2026, Cygnet Infotech acquired a majority stake in the company, combining Glib.ai's document processing capabilities with Cygnet's tax technology and fintech solutions for enterprise clients.

70%Turnaround time reduction at Shriram Finance
55 secSBI Life Insurance claims processing time
15 minKYC turnaround (down from 6 hours)
93%Data extraction accuracy across 2,500+ agents

Overview

Glib.ai was founded by Dr. Mohit Shah and Purav Parekh as a bootstrapped venture targeting BFSI document workflows rather than competing as a general-purpose intelligent document processing (IDP) platform. The company secured its first two clients in 2021, a leading education lender and a top-tier non-banking financial company (NBFC) in India, validating its vertical focus before seeking outside capital.

The Cygnet Infotech acquisition changes Glib.ai's competitive position materially. Cygnet brings decades of enterprise software experience in tax technology and fintech, giving Glib.ai distribution into larger enterprise accounts across BFSI, manufacturing, logistics, and supply chain. Both co-founders remain operational leads post-acquisition. As Niraj Hutheesing, Cygnet Infotech, stated: "We will be able to leverage Glib's hi-tech OCR reading capabilities for Bank Statement Analysis, Financial Statement Analysis and Invoice Analysis and combine it with our Tax Technology and Fintech solutions to bring the most advanced solutions for our global BFSI and other enterprise clients."

Dr. Mohit Shah described the rationale from Glib.ai's side: "The Cygnet Infotech brand and their decades of experience shall help us align our offerings to a larger enterprise audience and create a truly world class product suite."

The platform's technical differentiation sits in combining optical character recognition (OCR) with proprietary machine learning for fraud detection and compliance validation. This positions Glib.ai closer to risk-aware lending automation than to pure document extraction vendors.

Production deployments with measured outcomes

Glib.ai's documented deployments concentrate in Indian banking and insurance, where regulatory compliance and on-premise deployment are non-negotiable requirements.

At Shriram Finance, a major Indian NBFC processing auto and two-wheeler loans, Glib.ai reduced average document handling time from 25 minutes to under 3 minutes, a 70% turnaround reduction. The deployment runs across 2,500+ field agents through a digital portal, achieves 93% data extraction accuracy, and operates on-premise in compliance with Reserve Bank of India (RBI) guidelines.

A large international bank reduced KYC turnaround time from 6 hours to 15 minutes using Glib.ai's platform, with manual error rates dropping from 10% to below 1% and data extraction accuracy exceeding 95%. The system cross-references extracted data against government ID registries and credit bureaus automatically, with human-in-the-loop handling for subjective cases.

SBI Life Insurance processed over 70,000 motor insurance policies with accuracy above 90%, reduced manual effort by 85%, and cut claims processing turnaround to 55 seconds per claim. Ramakrishna Rao Dandanayakula, Assistant Vice President of Technology at SBI Life, noted: "Glib AI has been an outstanding partner for our Motor OCR needs. Their solution seamlessly handled a diverse set of over 70,000 insurance policies from multiple insurers, consistently delivering accuracy rates above 90%."

A leading Indian public sector bank uses Glib.ai to process 20 lakh (2 million) documents annually across retail and MSME lending, achieving a 70% reduction in processing time through straight-through processing (STP) with human-in-the-loop exception handling.

These outcomes are self-reported via Glib.ai's case study pages and carry no independent third-party verification. The consistency across deployments, 65-70% turnaround reductions and 90-95% extraction accuracy, aligns with published IDP benchmarks for BFSI workflows.

How Glib.ai handles BFSI document workflows

Glib.ai's pipeline combines OCR, natural language processing (NLP), and computer vision to classify, extract, and validate documents across the lending and insurance lifecycle. The platform supports bank statements, financial statements, income tax returns (ITR), GST returns, invoices, e-way bills, and KYC forms within a single deployment.

Fraud detection runs as an integrated layer rather than a post-processing step. The system identifies suspicious transaction patterns during extraction and cross-validates identity documents against external registries in real time. This design means a loan officer reviewing flagged documents sees both the extracted data and the fraud signal in the same interface, rather than reconciling outputs from separate systems.

The December 2024 launch of FinRay 2.0 extended this capability to multi-language financial statement analysis for lending workflows. Francesco Garcia, Director of Merchant Credit Risk at an unnamed firm, described the use case: "GLIB.ai will help us assess merchant risk in a better and more efficient manner. With FinRay, we can quickly analyze key ratios from financial statements across multiple languages."

Human-in-the-loop integration handles exception cases without breaking the automation pipeline. Staff review flagged documents through a dedicated interface, and their decisions feed back into the model for continuous learning. This approach lets institutions maintain straight-through processing rates for clean documents while keeping human judgment in the loop for edge cases.

Use cases

Banking and lending

Credit assessment uses automated bank statement analysis to generate risk scores for loan decisioning. KYC processing handles identity verification for customer onboarding with cross-validation against external databases. Document verification covers proof of address, income statements, ITR, and GST returns across retail and MSME lending portfolios.

Insurance claims

Claims processing automates death certificates, claim forms, and supporting documents. Motor insurance policy processing, as demonstrated in the SBI Life deployment, handles multi-insurer document sets with accuracy above 90%. Compliance documentation generates regulatory filing records and audit trails automatically.

Financial operations

Account reconciliation matches transactions and flags exceptions across accounts. Invoice management covers supplier validation and payment processing. FinRay 2.0 handles financial statement analysis across multiple languages for merchant credit risk assessment.

Technical specifications

Component Details
AI technologies OCR, NLP, computer vision, deep learning, proprietary ML algorithms
Document types Bank statements, invoices, identity documents, financial statements, ITR, GST returns, e-way bills, KYC forms
Processing volume 2 million+ documents annually in documented deployments
Extraction accuracy 93-95% in production deployments (vendor-reported)
Integration Core banking systems, insurance platforms, lending workflows
Compliance ISO 27001, SOC 2, GDPR, RBI guidelines
Deployment Cloud and on-premise; on-premise validated for RBI-regulated environments

Competitive position

Glib.ai competes in a segment where vertical depth matters more than breadth. Unlike horizontal IDP platforms that serve any document type across any industry, Glib.ai's entire product surface targets BFSI compliance requirements: RBI-compliant on-premise deployment, integrated fraud detection, and audit trails built for regulated lending and insurance workflows.

The Cygnet Infotech acquisition shifts the competitive calculus. Previously a bootstrapped startup competing against established vendors, Glib.ai now has access to Cygnet's enterprise distribution and fintech client base. The combined entity targets BFSI, manufacturing, logistics, and supply chain, though production evidence currently concentrates in banking and insurance.

For Indian financial institutions where on-premise deployment and RBI compliance are requirements, Glib.ai's documented outcomes and certification posture make it a credible shortlist candidate. For global enterprises or cloud-first deployments outside regulated Indian markets, the evidence base is thinner.

Resources