← Back to DTI Group's agentic AI platform COGNAiO cuts loan document processing time 80%

Source proof for DTI Group's agentic AI platform COGNAiO cuts loan document processing time 80%

Source-bound proof

Verified source excerpts for every supported field

Each colour maps a published value to the exact source passage used to support it. Only bounded excerpts are public; administrators can inspect the complete captured source.

12 fields supported

Title

DTI Group's agentic AI platform COGNAiO cuts loan document processing time 80%

derived · high
…ly, each requiring manual reconciliation across a wide range of document types. With COGNAiO, processing time per case was reduced by 80%, freeing staff to focus on higher-value tasks," recalls Manuel Gübeli, Solution Architect. These enhanced capabilities ch…

Description

DTI Group, a Switzerland-based document automation company, built COGNAiO, an agentic AI platform on Microsoft Azure and Microsoft Foundry (including Azure Document Intelligence, Azure Kubernetes Service, and Azure Database for PostgreSQL) that interprets, validates, and acts on information in documents, emails, images, and data feeds by meaning and context rather than templates. Users describe processes in natural language to deploy agents without code. One retail bank customer processing over 50,000 loan agreements annually reduced processing time per case by 80% using COGNAiO. In a high-volume banking deployment, the platform processed pages in an average of 2.5 seconds against a required service-level threshold of 6 to 9 seconds, with document understanding and validation accuracy approaching 99%.

derived · high
…g and context, validate information, and connect outputs to downstream systems. COGNAiO analyzes documents in seconds, with accuracy approaching 99%. In one loan-processing scenario, processing time was reduced by 80% per case, f…

Company

DTI Group

quote · high
…ed by 80% per case, freeing teams to focus on higher value work. READ THE STORY Switzerland-based DTI Group has decades of experience helping organizations manage complex, document-heavy processes. Since establishing a dedicated innovation hub in Germany in 2019, the company h…

Country

Switzerland

classification · high
…e Database for PostgreSQL SERVICES AND SUPPORT ORGANIZATION SIZE 1-49 employees COUNTRY/REGION Switzerland BUSINESS NEED AI and Agents INDUSTRY Software, Data and Platforms Share the sto…

Industry

Technology & Software

classification · medium
…TION SIZE 1-49 employees COUNTRY/REGION Switzerland BUSINESS NEED AI and Agents INDUSTRY Software, Data and Platforms Share the story Explore related stories /h2> "> Explore related stories Stories…

Problem

Document-heavy workflows such as customer onboarding, claims handling, and invoicing depended on manual review and reconciliation, slowing decisions, since rule-based automation could not reliably interpret unstructured inputs or connect different types of information within a single process.

derived · high
…and scanned documents. Traditional automation struggles to meet this challenge. Rule-based systems cannot reliably interpret unstructured inputs or connect different types of information within a single process. As volumes and complexity increase, outcomes slow and errors emerge. Yet with t…

Solution

DTI Group built COGNAiO on Microsoft Azure and Microsoft Foundry, including Azure Document Intelligence, Azure Kubernetes Service, and Azure Database for PostgreSQL; its agents interpret documents by meaning and context rather than templates, and users describe processes in natural language to deploy agents without code.

derived · high
…feeds, working together across steps to connect inputs with downstream actions. Built on Microsoft Azure and using capabilities in Microsoft Foundry (including Azure Document Intelligence ), Azure Kubernetes Service (AKS), and Azure Database for PostgreSQL , the platform combines advanced AI with secure, enterprise-grade infrastructure. Rather than relying on templates, COGNAiO understands documents by meaning and…

Business value

COGNAiO analyzes documents in seconds with accuracy approaching 99%; in a high-volume banking deployment, pages were processed in an average of 2.5 seconds against a required service-level threshold of 6 to 9 seconds; one retail bank customer processing over 50,000 loan agreements annually reduced processing time per case by 80%, freeing staff for higher-value tasks.

derived · high
…h accuracy levels approaching 99%, while maintaining high levels of automation. In a high-volume banking deployment, pages were processed in an average of 2.5 seconds, well below required service-level thresholds of 6 to 9 seconds. "One retail bank customer processed over 50,000 loan agreements annually, e…

AI capabilities

Document Intelligence, Agentic AI, Natural Language Processing

classification · high
…ned to bring agent-based document intelligence into complex workflows at scale. Its agents interpret, validate, and act on information in documents, emails, images, and data feeds, working together across steps to connect inputs with downstream actions. Built…

Technology

Microsoft Azure, Microsoft Foundry, Azure Document Intelligence, Azure Kubernetes Service, Azure Database for PostgreSQL

classification · high
…apka, Chief Product Officer, DTI Group Discover more details CUSTOMER DTI Group PRODUCTS Azure Microsoft Foundry Azure Document Intelligence in Foundry Tools Azure Kubernetes Service Azure Database for PostgreSQL SERVICES AND SUPPORT ORGANIZATION SIZE 1-49 employees COUNTRY/REGION Switzerlan…

Headline outcome

derived · high
…ly, each requiring manual reconciliation across a wide range of document types. With COGNAiO, processing time per case was reduced by 80%, freeing staff to focus on higher-value tasks," recalls Manuel Gübeli, Solution Architect. These enhanced capabilities ch…

Use case type

Document processing

classification · high
…polet, CEO, DTI Group Introducing document intelligence at scale The result was COGNAiO , an AI-native platform designed to bring agent-based document intelligence into complex workflows at scale. Its agents interpret, validate, and act on information in documents, emails, i…
Capture details
Captured
15 Sept 2026, 08:17 UTC
Extractor
fetch-strip@1
Snapshot hash
4b93263cf7122b7ef602e9e0795ef05d57562640f30dbbffb90fb205d6848e73