Normalized claim
Suspicious transaction report processing speedup: 3 x increase
validate and submit anti-money laundering suspicious-transaction reports three times faster
KOHO — Kortex AI for secure generative AI across fintech workflows
Use case typeCode assistantUpdated Jun 13, 2026
KOHO, a Canadian fintech, built Kortex AI to embed generative AI across internal workflows while maintaining strict data privacy and governance. The solution gives employees and developers secure access to large language models for research, coding assistance, meeting preparation, regulatory compliance, and operational analysis. KOHO uses the platform to improve productivity, accelerate suspicious transaction report processing, and speed up security operations.
Reported outcomes
3x
Suspicious transaction report processing speedupTime & speed
1 daysMoney-laundering investigation time reduction−50%Security event resolution time reduction120-140%Pull request throughput increase+66%Deployment frequency increase2 daysFeature delivery time
Strategic outcomes
New product / capabilityBuilt secure enterprise AI platformNew product / capabilityEnabled secure LLM access across workflowsSpeed & agilityAccelerated suspicious transaction processingRisk & complianceMaintained strict data privacy and governance
Catalog median for time & speed deployments: +60% across 143 reported metrics. Compare benchmarks →
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Money-laundering investigation time reduction: 1 days decrease
reducing money laundering investigation time by days
Normalized claim
Security event resolution time reduction: 50% decrease
Security event resolution time has dropped 50 percent
Normalized claim
Pull request throughput increase: 120-140% increase
with 120-140 percent higher pull request throughput
Normalized claim
Deployment frequency increase: 66% increase
and 66 percent more frequent deployments
Normalized claim
Feature delivery time: 2 days decrease
the enhancement was built in 2 days
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- KOHO
- Provider
- AWS
- Maturity
- Production
- Linked source
- AWS Customer Case Study
The solution gives employees and developers secure access to large language models for research, coding assistance, meeting preparation, regulatory compliance, and operational analysis
Customer identity supportedSource describes one deploymentMaturity supported
Primary read
Use case focus
Showing 3 of 5
- 1Enterprise AI platform
- 2Secure copilot-style assistant
- 3RAG-enabled internal search
- KOHO wanted to move beyond isolated AI tools and embed generative AI across workflows.
- The company needed secure access to LLMs without compromising financial data privacy, logging, or governance requirements.
- It also needed to support multiple business functions, including development, compliance, and operations, on a scalable AWS foundation.
- KOHO built Kortex AI on Amazon Bedrock to provide secure, scalable access to generative AI across the organization.
- The solution uses Amazon Aurora as a vector database to support retrieval augmented generation use cases.
- Amazon EKS provides container orchestration for AI workflows, APIs, and interfaces.
- KOHO added centralized authentication, logging, model approval lists, and restricted data-source access to meet financial-services controls.
- Suspicious transaction report processing became 3x faster.
- Money-laundering investigation time was reduced by days.
- Security event resolution time dropped 50%.
- Pull request throughput increased by 120-140%.
- Deployments became 66% more frequent, and a requested feature was built in 2 days.
Architecture
KOHO built Kortex AI on Amazon Bedrock to provide secure access to LLMs within AWS. Amazon Aurora supports vector storage for RAG. Amazon EKS orchestrates AI workflows, API interactions, and user interfaces. The platform includes centralized authentication and logging, model approval lists, and restricted source access to maintain financial-services governance and privacy controls. KOHO also plans to use Amazon Bedrock AgentCore for future multistep agent workflows.
Sources & evidence1
Evidence: Medium65/100Evidence strength
- Customer explicitly identified
- Deployment status explicitly supported
- Primary source available
- Quantified outcome available
- Technical implementation details available
Type: Customer StoryPublished: Jun 8, 2026Publisher: AWSEvidence: PrimaryConfidence: High
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
AI patterns
Browse the catalog
Was this useful?
Community
Comments
No published comments yet.