Bankdata pilots Microsoft's agentic COBOL Migration Factory against its 70-million-line mainframe estate
Bankdata· DenmarkMicrosoft Semantic Kernel · GPT-4.1 · AutoGen +1
Bankdata, a technology company owned by a consortium of Danish banks representing over 30% of the Danish banking market, has over 70 million lines of code still running on the mainframe. Bankdata worked with Microsoft to build the COBOL Agentic Migration Factory (CAMF), a multi-agent system built on Semantic Kernel with specialized agents (COBOLAnalyzerAgent, JavaConverterAgent, DependencyMapperAgent) that analyze, map dependencies for, and convert COBOL programs into modern Java Quarkus applications. Microsoft tested the framework on a small COBOL module donated by Bankdata, progressively validating it on more complex COBOL call chains reaching a call-chain depth of level 3 as the tool matured into its current multi-agent form.
Overview
Bankdata, a technology company owned by a consortium of Danish banks representing over 30% of the Danish banking market, has over 70 million lines of code still running on the mainframe. Bankdata worked with Microsoft to build the COBOL Agentic Migration Factory (CAMF), a multi-agent system built on Semantic Kernel with specialized agents (COBOLAnalyzerAgent, JavaConverterAgent, DependencyMapperAgent) that analyze, map dependencies for, and convert COBOL programs into modern Java Quarkus applications. Microsoft tested the framework on a small COBOL module donated by Bankdata, progressively validating it on more complex COBOL call chains reaching a call-chain depth of level 3 as the tool matured into its current multi-agent form.
The challenge
Bankdata, a technology company owned by a consortium of Danish banks representing over 30% of the Danish banking market, has over 70 million lines of code still running on the mainframe. Modernizing away from the mainframe is complex because many COBOL modules are tied to non-functional behaviors such as batch throughput, I/O handling, JCL orchestration and strict SLAs, and re-platforming has historically been a manual, time-consuming and costly affair, compounded by a shrinking pool of COBOL experts and loss of institutional knowledge.
The solution
Microsoft, in collaboration with Bankdata, built the COBOL Agentic Migration Factory (CAMF), a multi-agent system orchestrated via Microsoft Semantic Kernel with specialized agents — COBOLAnalyzerAgent, which scans COBOL files to understand their structure and logic; JavaConverterAgent, which converts COBOL source code to Java Quarkus; and DependencyMapperAgent, which analyzes copybook dependencies and generates diagrams — that automates analysis, transformation, dependency mapping and reporting for migrating COBOL programs into modern Java Quarkus applications.
Sources
Open any source and check the claim yourself — that is the point of the register.
This record was researched and written with AI assistance, and its claims were checked against the sources above. (EU AI Act art. 50 transparency notice.)
Other financial services entries in the register.
Navy Federal Transforms Service With AI
Navy Federal Credit Union is reshaping banking for military members by unifying data and leveraging generative and agentic AI on the Databricks Data + AI Platform. By embracing AI-augmented workflows and upskilling teams, Navy Federal delivers customized services while streamlining productivity through responsible change management and data readiness.
Navy Federal Credit UnionDatabricks
Lloyds Banking Group cuts mortgage income verification from days to seconds with ML
Lloyds Banking Group, the UK's largest digital bank, migrated 15 modelling systems from on-premise infrastructure to Google Cloud's Vertex AI (now Agent Platform), giving over 300 data scientists and AI developers scalable machine learning capabilities. In six months the bank ran 80 new ML experiments and launched 18+ GenAI systems into production, including an algorithm that reduces the income verification step in mortgage applications from days to seconds. The migration also cut unplanned ML platform downtime to zero and saved 27 CO2 tonnes of operational emissions.
Banking Innovator bunq Supports Growth, Strengthens Security Using AWS
bunq, a Dutch neobank with over 11 million users across Europe, uses Amazon Bedrock for several generative AI use cases including summarizing new user data with large language models, removing the need for agents to process onboarding documents manually. Using Amazon Bedrock, bunq tripled user support process efficiency while maintaining over 90 percent accuracy. Sensitive data stays within bunq's AWS virtual private cloud, supporting GDPR and PCI DSS compliance alongside tools such as AWS CloudHSM, AWS Security Hub and AWS KMS.
bunq· NetherlandsAmazon Bedrock · AWS CloudHSM · AWS Security Hub +1TBC Bank Operationalizes Trusted Data with Lakebase
TBC Bank, the largest banking group in the Caucasus region, built a Lakehouse on Databricks and adopted Lakebase and Databricks Apps to move from on-premises SQL Server instances and month-long reporting cycles to self-service analytics and AI-driven applications, including a web-based AI chatbot and AutoML-based credit risk scoring. Credit risk model deployment fell from 14 weeks to two days, and more than 600 users regularly query governed data through Genie.
TBC Bank· GeorgiaDatabricks Platform · Lakebase · Databricks Apps +4Was this helpful?
Your feedback helps us improve our use case database