How to Modernize Core Banking Systems

In this Deep Dive edition of Fintech Wrap Up, I explored the complex world of modernizing core banking systems—and why it’s becoming more critical than ever, based on the experience of Tuum. Banks today are grappling with mounting economic pressures, outdated technology, and fierce competition. With cost-to-income ratios hovering around 60%, there’s a growing urgency to rethink operations and core infrastructure. What really stood out is how top-performing banks—the true tech leaders—are cracking the code by shifting the bulk of their IT budgets toward innovation instead of just keeping legacy systems on life support. These forward-thinking players are allocating up to 75% of their tech spend to “build the bank” initiatives, allowing them to launch products faster, enhance customer experiences, and operate far more efficiently. One of the biggest questions banks face is how to actually approach modernization. Full core replacements are expensive and high-risk, while greenfield builds don’t always solve existing legacy issues. That’s where Progressive Migration comes in—it offers a middle path, allowing banks to transition gradually off outdated systems without disrupting business as usual. A real-life case study from Tuum brings this approach to life. It shows how a mid-sized bank migrated accounts, lending, and reporting to a modern core platform—all while maintaining service continuity. By leveraging smart APIs and data lakes, they not only stayed on budget but also automated key processes and eliminated years of technical debt. If you’re navigating a digital transformation—or just curious about the future of core banking—this one’s packed with insight. 🚀 #banking #fintech #corebanking

𝐌𝐨𝐬𝐭 𝐛𝐚𝐧𝐤𝐬 𝐚𝐫𝐞𝐧’𝐭 𝐬𝐭𝐫𝐮𝐠𝐠𝐥𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝐀𝐈. 𝐓𝐡𝐞𝐲’𝐫𝐞 𝐬𝐭𝐫𝐮𝐠𝐠𝐥𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝐂𝐎𝐁𝐎𝐋. This year, Morgan Stanley quietly did something bold. They built DevGen(.)AI a GPT-based tool trained not on GitHub, but on their own legacy code: COBOL, JCL, SAS, in-house Perl scripts. And in just a few months: ✔ 9 million lines of legacy code processed ✔ 280,000 developer hours saved ✔ 15,000+ engineers using it globally This isn’t about generating new code. It’s about making old code readable, documenting logic buried in 40-year-old systems so modern developers can rewrite it in Python or Java. Why it matters: Most AI copilots can’t help here. Legacy logic doesn’t live on the internet. It lives in ancient batch jobs, undocumented macros, and formats no modern LLM was trained on. Morgan Stanley’s edge? They fine-tuned the model on proprietary systems. Now they’re getting cleaner outputs, faster onboarding, and tighter governance, with no hallucinations. Meanwhile, off-the-shelf tools struggle with context, privacy, and legacy syntax. 𝐓𝐡𝐢𝐬 𝐢𝐬 𝐰𝐡𝐞𝐫𝐞 𝐆𝐞𝐧𝐀𝐈 𝐦𝐨𝐯𝐞𝐬 𝐟𝐫𝐨𝐦 𝐚𝐬𝐬𝐢𝐬𝐭𝐚𝐧𝐭 𝐭𝐨 𝐬𝐲𝐬𝐭𝐞𝐦𝐬 𝐚𝐫𝐜𝐡𝐚𝐞𝐨𝐥𝐨𝐠𝐢𝐬𝐭. The economics are compelling too. At ~$100/hour, those 280,000 saved hours equal $28 million unlocked. The payback period? Less than 24 months. Here’s the real strategy shift: Modernization is not a side project. It’s been a board-level priority for years now. But now, something meaningful can be done. 60 - 80% of IT budgets go to maintenance And COBOL talent is vanishing (if not already) If you’re not using AI to decode your own systems by 2025, your risk isn’t just technical. It’s institutional memory loss. Your codebase is your architecture. Your constraints. Your truth. Modernization isn't about speed. It's about clarity. 𝐍𝐨𝐭 𝐚 𝐌𝐨𝐫𝐠𝐚𝐧 𝐒𝐭𝐚𝐧𝐥𝐞𝐲? You don’t need to be. Here’s what smaller banks and tech teams can do: → Start with documentation, not translation. Use AI to generate English summaries of your core legacy apps first. Focus on clarity, not code conversion. Tools like GPT-4 can already help here without full custom training. → Fine-tune with what you own. If your codebase is too proprietary for public copilots, use small internal LLMs or embeddings over your repos. Even a basic RAG setup over COBOL comments can lift onboarding speed. → Prioritize by exposure, not convenience. Don’t modernize what’s easiest. Modernize what’s riskiest. Start with systems that touch audit, risk, or customer data. → Invest in “translators,” not just devs. Your most critical hires aren’t Python experts. They’re the ones who can bridge legacy logic and modern architecture. Pair them with AI and scale their impact. The goal isn’t to match Morgan Stanley’s throughput. The goal is to stop bleeding institutional knowledge with every retirement. If your AI roadmap doesn’t include your oldest systems, you’re modernizing the front while the foundation crumbles

Banks’ biggest tech challenge isn’t upgrading legacy systems- it’s integrating an entirely new (Gen)AI layer with orchestration in the lead. And making it work across functions. Too many banks often start with the wrong focus. Whereas dealing with legacy infrastructure is inevitable, it can become a blind spot without the right understanding of what it needs to achieve. Delivering agile, intelligent services that anticipate customer needs should be the goal. Here is a high-level overview of how the back end can be adjusted: 𝟭. 𝗢𝗿𝗰𝗵𝗲𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻 𝗲𝗻𝗴𝗶𝗻𝗲: -   An orchestration layer sits atop core systems, routing everything - from customer questions to fraud alerts - to the right AI service. -   Modern APIs abstract legacy systems into modular services, so AI features can be added or swapped without changing existing workflows. 𝟮. 𝗥𝗲𝗮𝗹-𝘁𝗶𝗺𝗲 𝗶𝗻𝗳𝗿𝗮𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲: -   Real-time data feeds stream transactions, balance changes and logins as they happen. -   A unified data hub brings together customer details, activity patterns and risk ratings so every AI tool works from the same information. 𝟯. 𝗗𝗮𝘁𝗮-𝗱𝗿𝗶𝘃𝗲𝗻 𝗶𝗻𝘀𝗶𝗴𝗵𝘁𝘀: -   Requests are automatically enriched with live account balances, recent transactions and open support tickets - ensuring the AI’s output reflects up-to-date information. -   Data is fetched on demand from indexed records, so the AI stays current without the expense of retraining the entire model for every update. 𝟰. 𝗦𝗲𝗰𝘂𝗿𝗶𝘁𝘆 & 𝗴𝗼𝘃𝗲𝗿𝗻𝗮𝗻𝗰𝗲: -   Data stays encrypted end-to-end, from intake to AI output. -   Automated audits flag bias and log every decision. -   Failure simulations uncover hidden risks before they impact customers. 𝟱. 𝗠𝗼𝗱𝘂𝗹𝗮𝗿 𝘀𝗲𝘁-𝘂𝗽: -   Modern interfaces turn core banking, payment and CRM systems into plug-and-play modules. -   Behind the scenes, back-end services can be updated piece by piece without interrupting the AI layer. 𝟲. 𝗜𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗲𝗱 𝗱𝗲𝗹𝗶𝘃𝗲𝗿𝘆 𝘁𝗲𝗮𝗺𝘀: -   Small, cross-functional teams manage everything from data ingestion to model deployment and monitoring. -   Defined roles and fast feedback loops keep projects compliant and focused on real customer needs. The GenAI layer doesn’t just sit on top of the existing setup – it’s a complete overhaul of the tech architecture and the business logic behind it. Opinions: my own, Graphic source: BCG 𝐒𝐮𝐛𝐬𝐜𝐫𝐢𝐛𝐞 𝐭𝐨 𝐦𝐲 𝐧𝐞𝐰𝐬𝐥𝐞𝐭𝐭𝐞𝐫: https://lnkd.in/dkqhnxdg

For Banks, (Gen)AI Tech Architecture Requires New Capabilities 💡 Put AI at the center of tech and data. Making AI work at scale requires rethinking the architecture itself. This demands changes across tech, data, and infrastructure: 🌐 Workflow integration requires deep orchestration. As banks evolve their AI capabilities, the challenge has shifted from developing specialized models to integrating them intelligently. Orchestration matters, and GenAI makes this nonnegotiable. Banks must design routing mechanisms that direct specific information to the best-fit model while also integrating proprietary data through techniques like retrievalaugmented generation (RAG) and domain-specific small language models (SLMs). Orchestration will become even more critical as agentic AI use expands so that banks can coordinate decision execution as well as information flows. But as financial institutions develop increasingly complex ecosystems, banks will need holistic oversight. ☁️ Data availability, not just accuracy, defines AI performance. Most AI failures in banking aren’t about the models—they’re about slow, incomplete, or fragmented data. Unlocking AI’s full potential requires addressing outdated systems and IT shortcuts, setting up strong governance, and enabling efficient data integration across cloud and on-premise environments. LLMs will take a central role in banking AI, but they won’t be sufficient. Many financial tasks are simply too specialized to rely on broad, general-purpose models, even when these are customized for particular domains. 👨💻 Core layers must modernize. Most banking systems are a technological patchwork that obstructs the dynamic, real-time, and unstructured capabilities essential for innovative AI applications. Simply adding AI components to existing infrastructure won’t work. Leading institutions are demonstrating a new approach. Commonwealth Bank of Australia has implemented an event-driven architecture and an AI-powered transaction core. These allow for real-time fraud detection and response, contributing to a 50% drop in scam losses and a 30% decrease in customer-reported fraud. 🤖 Hybrid infrastructure is essential. Today, AI systems can flag risks, surface insights, and suggest pricing changes—but most don’t trigger real-time adjustments. This must change. There are many opportunities where predictive and agentic AI can work together to propose an action and then implement it without exposing the bank to risk. For these opportunities to expand, infrastructure needs to be hybrid. It must cut across on-premise, cloud, and edge environments to enable high degrees of modularity and the widespread use of application programming interfaces and micro-services. Source: Boston Consulting Group (BCG) - https://shorturl.at/fiSpV #Innovation #Fintech #Banking #FinancialServices #AI #MachineLearning #Data #Cloud #LLMs #GenAI #AgenticAI

Confronting Technical Debt: The Strategic Imperative for Banking IT Over the last 30 years, I have seen banking software where only 2% of the code is still active. A resulting mantra in many banks is: "never touch a running system." However, the accelerating increase in regulatory requirements and innovation forces banks to change faster and faster. As a consequence, technical debt in banking IT is a ticking time bomb that threatens the stability and future of the entire industry. Decades of neglect, quick fixes, and prioritization of short-term gains have left banks saddled with creaking legacy systems that are ill-equipped to handle the demands of the digital age. This accumulated debt is not just an IT problem; it's a strategic risk that can lead to catastrophic consequences. Security breaches, system outages, and missed opportunities for innovation are just some of the dangers that lurk beneath the surface. Banks must confront this debt head-on, investing in modernization and adopting agile practices to build a technology infrastructure that is resilient, secure, and adaptable. Failure to do so will leave them vulnerable in an increasingly competitive and technologically driven landscape, risking irrelevance and ultimately, extinction. In this labyrinthine world of banking IT, I see software refactoring as the unsung hero battling the looming specter of technical debt. It's a meticulous process of restructuring existing code, not to add new features, but to enhance its internal structure and maintainability. Refactoring breathes new life into aging systems, making them more adaptable to evolving business needs and regulatory requirements. It's a proactive measure, akin to regular maintenance of a complex machine, that prevents the accumulation of technical debt and its associated risks. By improving code readability, reducing complexity, and eliminating redundancies, refactoring enhances the efficiency and reliability of banking software. My belief and recommendation: A bank should spend 20% of its software development budget on refactoring and should prefer vendor software with a transparent and relevant refactoring approach. It's an investment in the future, ensuring that the technology infrastructure remains robust, secure, and capable of supporting innovation. In the long run, refactoring is not just a technical necessity but a strategic imperative for banks aiming to thrive in the digital age. #banking #IT #digital #technicaldebt #refactoring #SundayThoughts

𝐂𝐨𝐫𝐞 𝐁𝐚𝐧𝐤𝐢𝐧𝐠 & 𝐏𝐚𝐲𝐦𝐞𝐧𝐭 𝐇𝐮𝐛 𝐓𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐚𝐭𝐢𝐨𝐧 — 𝐰𝐡𝐲 𝐚𝐫𝐞 𝐬𝐨 𝐦𝐚𝐧𝐲 𝐛𝐚𝐧𝐤𝐬 𝐦𝐨𝐯𝐢𝐧𝐠 𝐭𝐡𝐢𝐬 𝐰𝐚𝐲? “Payment hubs” as a concept have been around for years. But in many banks the reality is still core-centric: every new rail, channel or regulation ends up as another change in the core. At some point the core stops being an enabler and becomes the bottleneck. 🔴 𝐂𝐨𝐫𝐞-𝐂𝐞𝐧𝐭𝐫𝐢𝐜 (𝐓𝐨𝐝𝐚𝐲) ▪ Channels and rails are hard-wired into the core ▪ Payment logic sits inside the core: routing, rules, limits, pricing, file formats, fraud/AML, etc. ▪ Every scheme / product / compliance change = a core project ▪ Long release cycles, heavy testing, high delivery risk Over time the architecture looks less like a design – and more like a history of quick fixes. 🟢 𝐇𝐮𝐛-𝐂𝐞𝐧𝐭𝐫𝐢𝐜 (𝐓𝐨𝐦𝐨𝐫𝐫𝐨𝐰) A modern payment hub introduces a single orchestration layer between channels/rails and the core: ▪ Channels and schemes integrate once – into the hub ▪ The hub owns payment logic: routing, rules engine, limits, dynamic fees, ISO 20022 / file transforms, enrichment ▪ Fraud, AML and monitoring are centralised around the payment flows ▪ The core becomes a stable ledger for accounts, loans, limits and posting Adding a new rail or use case is no longer core surgery – it’s hub configuration and integration. 📈 𝐖𝐡𝐚𝐭 𝐭𝐡𝐢𝐬 𝐭𝐲𝐩𝐢𝐜𝐚𝐥𝐥𝐲 𝐝𝐞𝐥𝐢𝐯𝐞𝐫𝐬 ▪ Faster delivery – weeks instead of months for many changes ▪ Less dependency on core releases and vendor roadmaps ▪ One integration point for channels and schemes ▪ Lower operational risk and more room for innovation on top of a stable core ⚠️ 𝐁𝐚𝐧𝐤𝐬 𝐭𝐡𝐚𝐭 𝐬𝐭𝐢𝐥𝐥 𝐫𝐮𝐧 𝐩𝐚𝐲𝐦𝐞𝐧𝐭𝐬 𝐬𝐨𝐥𝐞𝐥𝐲 𝐨𝐧 𝐭𝐡𝐞 𝐜𝐨𝐫𝐞 𝐚𝐫𝐞 𝐠𝐫𝐚𝐝𝐮𝐚𝐥𝐥𝐲 𝐟𝐚𝐥𝐥𝐢𝐧𝐠 𝐛𝐞𝐡𝐢𝐧𝐝 𝐭𝐡𝐨𝐬𝐞 𝐦𝐨𝐯𝐢𝐧𝐠 𝐭𝐨 𝐡𝐮𝐛-𝐜𝐞𝐧𝐭𝐫𝐢𝐜 𝐚𝐫𝐜𝐡𝐢𝐭𝐞𝐜𝐭𝐮𝐫𝐞.

Banks are about to waste billions replacing their COBOL systems. Most think modernization means ripping everything out and starting fresh. But here's what nobody's talking about... COBOL is the big calculator that nobody hears about in As400 or mainframe. And the banks who tried replacing it? → 70% of projects failed → Average cost: $500M+ in losses → Years of delays and setbacks The smartest banks are doing something different... They're integrating modern APIs with existing COBOL. Building bridges instead of demolishing foundations. Major companies saved $300M using this approach. Their secret? 1. Keep core COBOL systems intact 2. Add modern interfaces on top 3. Gradually enhance vs. replace Sometimes the best innovation isn't starting over. It's building on what already works.

👀 I’ve been reviewing the latest architecture for Infosys Finacle on IBM Power11, and I’m impressed by how it addresses the "stability vs. agility" trade-off we often see in core banking. For those working with mission-critical core banking applications, the goal is always to modernize without introducing risk. What I find most compelling about this new Power11-based solution is its practical approach to the hybrid cloud. It allows you to keep your core database and applications on highly secure AIX partitions while simultaneously running digital and AI suites in containers on the same hardware. A few innovations in Power11 that makes this possible: 🔂 Reliability: The promise of zero planned downtime through Live Partition Migration (LPM) is key for maintaining 24/7 banking operations. 🔐 Future-Ready Security: It’s great to see "quantum-safe" cryptography integrated at the hardware level—a necessary move as we look toward the next decade of compliance. 💰 Operational Efficiency: Using PowerVM for hardware-level virtualization means lower overhead and better performance for *Oracle-heavy core workloads* compared to traditional software hypervisors. This is a solution that offers great innovation while remaining so grounded in the realities of core banking. If you’re looking for a low-risk, high-performance path to modernization, this is a development worth your time. Take a look at the full breakdown of the solution paper written by Rebecca Gott and K.R. Venkatraman : https://lnkd.in/gzzDK4wy #Finacle #IBMPower11 #CoreBanking #DigitalTransformation #BankingTechnology #HybridCloud #FinTech

Community Banks and Credit Unions: Customers want to open an account in minutes. Your legacy workflow makes it feel like days. Most banks and CUs are still running legacy workflows originally built for in-branch processes. If your deposit account opening journey feels clunky, it’s a  for sure growth killer. So how do you fix it without blowing everything up? Here are 4 steps to modernize your deposit onboarding flow and  grow deposits in the process. 1️⃣ Map every step of your current process. You can’t modernize what you don’t fully understand. Start by creating a complete visual map of your deposit account opening workflow, including: • Online and in-branch steps • Internal handoffs and approvals • Manual reviews, verifications, or delays Then identify steps that can be streamlined, automated, or eliminated entirely. 2️⃣ Audit where applicants are dropping off. Most banks lose customers during onboarding and don’t know where or why. • Where do people abandon the form? • How long does it take to open an account? • Is there a spike in support tickets or failed verifications? Use analytics, session replays, and customer feedback to find the leaks. Fixing just one friction point could recover thousands in lost deposits. 3️⃣ Design with customers (and employees) in mind. Great onboarding is intuitive for the customer 𝘢𝘯𝘥 easy on your team. Make sure your redesign includes: • ID verification that works across devices • Clear status updates and progress bars • Fewer manual steps for back-office staff • Built-in compliance, not bolted on    If your workflow upgrades CX but overloads your staff, you’re just moving the problem downstream. 4️⃣ Pilot, measure, and optimize before scaling. Don’t launch a shiny new experience and walk away. Start with a measured rollout and track: • Completion rates • Account funding time • Cross-sell or deposit growth Then optimize before going wide. Use feedback loops to improve continuously. At Candescent, we help banks modernize deposit account opening so it's fast for customers and smooth for employee, all without adding manual inefficiencies behind the scenes that make your employees grumble. Want help talking through an improved process? Connect and shoot me a message and we'll get the conversation rolling.