National Australia Bank accelerates legacy migrations with Cursor

2026-04-23 · Source: Cursor Blog · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning, FinTech & Digital Financial Services · Depth: Intermediate, short

Summary

National Australia Bank (NAB) has standardized 6,000 developers on Cursor, an AI coding assistant, after evaluating alternatives like Amazon Q and GitHub Copilot. This adoption has led to a threefold acceleration in legacy modernization projects, including refactoring a Silverlight/.NET monolith to Java microservices and migrating Assembly-based mainframe applications. Additionally, NAB's merchant services team developed a new hardware-agnostic payment application in three weeks, significantly faster than the original four-month estimate. NAB chose Cursor for its model flexibility, superior codebase understanding across diverse tech stacks (Java, React, COBOL, Assembly), and extensibility through an internal context engineering library (NAB CEL). The bank plans to expand Cursor's use to over 10,000 employees across various functions and integrate it into the entire software development lifecycle, including code review and QA.

Key takeaway

For CTOs and VPs of Engineering evaluating AI coding assistants for large-scale enterprise adoption, consider Cursor's demonstrated ability to accelerate legacy migrations and greenfield development by 3x or more. Your teams can achieve significant productivity gains and tackle previously unfeasible projects by prioritizing solutions offering model flexibility, deep codebase understanding across diverse tech stacks, and extensibility for custom guardrails and shared knowledge.

Key insights

Enterprise-wide AI coding assistant adoption can significantly accelerate legacy modernization and greenfield development.

Principles

• Model flexibility optimizes cost/latency for varied tasks.
• Deep codebase understanding is critical for complex migrations.

Method

NAB used Cursor's Ask Mode for business logic visualization, Plan Mode with custom skills for user stories and API specs, and Composer/Opus models for implementation, enabling rapid pre-development and execution.

In practice

• Refactor monoliths to microservices 3x faster.
• Migrate mainframe Assembly code with AI assistance.
• Build greenfield apps in weeks, not months.

Topics

• Cursor AI Platform
• Legacy Modernization
• Mainframe Migration
• Microservices Architecture
• Enterprise AI Adoption

Best for: CTO, VP of Engineering/Data, Executive, AI Engineer, Software Engineer, Director of AI/ML

Related on AIssential

• Money Forward brings Cursor’s coding agents to product, design, and QA — AI coding agents can significantly boost productivity across an entire software development lifecycle, not just engineering.
• Sources: Cursor in talks to raise $2B+ at $50B valuation as enterprise growth surges — AI coding startup Cursor is securing significant funding and revenue growth despite competition, driven by strategic model diversification.
• Meet the new Cursor — Cursor 3 unifies AI agent-driven software development in a multi-workspace environment.
• New in Migrations: Faster and More Predictable — Lakebridge automates data warehouse migrations using AI-powered SQL conversion and comprehensive environment assessment.
• Why Cursor is the Enterprise AI Darkhorse of Generative AI — Cursor's "agent-first" interface and multi-agent parallelization redefine AI-augmented software development and knowledge work.
• Cursor's Wild Trajectory to being a Vibe Working Leader — AI platforms with integrated plugins are transforming work by enabling autonomous, multi-tool workflow automation.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Cursor Blog.