Modernizing Legacy Systems

Translating code like #COBOL or #RPG is not a modernization solution in itself, because although AI-powered tools like Ozgar.ai , #IBMBob and X-Analysis offer huge promise, it is only a starting point, not the destination. Moreover, moving to the cloud may be detrimental to modernization, as much of what makes #IBMi and #IBMz systems so special is embedded within the platform. Peerless levels of reliability, security and performance have been built and maintained through decades of optimization, performance tuning and tight coupling between software and hardware. This kind of intel is hard to replace.     Modernization of IBM platforms is a multi-dimensional problem that requires resolving disparate challenges such as on-premises dependencies, encryption, transactional integrity, security, system-level engineering, data residency, database architecture, scaling and disaster recovery. Translating code is only part of the story. How #IBM customers solve these issues whilst successfully integrating with everything around their mission-critical, core system is where the real modernization journey begins. 

Banks can’t afford to miss this one. Core banking modernization has long dominated banks’ boardrooms. AI is now flipping the script as it introduces - for the first time in decades - a new operating logic. Here are their options. For years, core transformation has been treated as a slow, multi-year journey - a balancing act between stability and change. Most banks took a cautious, incremental approach: patch the legacy, layer on digital, expose some APIs, and keep moving. It was less about reinvention - and more about risk containment. But this time is different. AI isn’t just another tool to plug into aging infrastructure. It demands - and enables - a fundamentally new way of operating. It runs on real-time data, modular architecture, and continuous orchestration. Traditional core systems weren’t designed for this. They operate in batch cycles, live in product silos, and require months to adapt to even small changes. Simply adding AI into that environment will not work. 𝗕𝗮𝗻𝗸𝘀 𝗻𝗼𝘄 𝗳𝗮𝗰𝗲 𝘁𝗵𝗿𝗲𝗲 𝗯𝗿𝗼𝗮𝗱 𝗰𝗵𝗼𝗶𝗰𝗲𝘀: 1. Keep modernizing around legacy cores - slow, expensive, and increasingly misaligned with business needs. 2. Attempt a full core replacement - a bold move that requires significant investment and careful execution. 3.  Rethink the role of the core entirely - and start decoupling intelligence, orchestration, and engagement from legacy constraints. That third path is fast gaining traction. Backbase is one of the leading examples. They’ve recently launched the world’s first AI-powered Banking Platform, built not as a replacement for the core, but as a new operating system around it. Here’s the set-up: 1. Unified Digital Banking Fabric – orchestrates onboarding, servicing, lending, investing, and more, across all channels. 2. Intelligence Fabric – embeds AI services, agentic automation, predictive models, and responsible AI governance into every layer. 3. Integration Fabric – connects seamlessly to existing cores, CRMs, fintechs, and third-party systems with 60+ pre-built connectors. 4. Composable Architecture – modular, cloud-native, and flexible – banks pick what they need, when they need it, without disruption. 5. Core-Agnostic: It wraps around legacy systems, accelerating transformation without the risks and costs of core replacement. This isn’t a full core replacement - it’s a strategic rearchitecture. It allows banks to keep what’s stable in the back, while radically upgrading the front and middle with intelligence, flexibility, and speed. It's a modular, composable approach that reflects the reality most banks face: evolve without breaking everything. AI won’t wait for five-year transformation roadmaps. It’s already reshaping the game - and the banks that win will be the ones building real capabilities, redesigning processes, and deploying at speed.   Opinions: my own, Graphic source: Backbase 𝐒𝐮𝐛𝐬𝐜𝐫𝐢𝐛𝐞 𝐭𝐨 𝐦𝐲 𝐧𝐞𝐰𝐬𝐥𝐞𝐭𝐭𝐞𝐫: https://lnkd.in/dkqhnxdg 

AI field note: Modernization is one of the most underappreciated forces for innovation (Southwest Airlines shows us why). When legacy systems finally get updated, two big things happen: 1️⃣ You can start improving services that were effectively frozen in time. 2️⃣ The cost and complexity of running those services drops—freeing up time, money, and focus for what’s next. But for a long time, modernization just wasn’t worth it. The juice wasn’t worth the squeeze. Projects kicked off with long planning cycles, manual analysis, and a lot of upfront investment—often without a clear path to value. That’s starting to change. AI is shifting what’s possible. It can help teams understand legacy code faster, accelerate planning, and reduce the rework that usually slows things down. With that, modernization becomes more viable, more targeted, and more focused on outcomes. It’s not just about updating systems—it’s about unlocking capacity, reducing friction, and making space for the next wave of innovation. Take Southwest Airlines. They needed to modernize their crew leave management system—a critical platform for scheduling, time off, and operations. Over time, the system had become harder to update. Technical debt made it difficult to plan changes, and documentation was limited. Each update required hours of manual analysis just to understand what the system was doing—slowing delivery and tying up valuable resources. But the pressure to modernize was growing. As operations evolved and employee needs changed, the system needed to be more flexible, more reliable, and easier to maintain. PwC partnered with Southwest to take a different approach. Using GenAI, we analyzed the legacy code and generated user stories—effectively mapping the system’s behavior and identifying what needed to change. That work: ⚡️ Cut backlog creation time by 50% 🌟 Produced user stories accepted 90% of the time without major rework 💫 Freed up 200+ hours across teams More importantly, it gave the team clarity and momentum—turning a slow, manual planning process into a faster, more focused path forward. Less time untangling the past. More time building what’s next—for their teams and their travelers. There’s never been a better time to modernize.

How Monzo Bank, Revolut, and Chime are changing banking Old banks are built on old tech. Digital banks are not. That’s the main difference. Instead of using one big, slow system, digital banks use many smaller systems connected by APIs. This makes apps faster, easier to update, and nicer to use. According to McKinsey, about 70% of banks now focus on cloud-based systems to grow faster and launch new products more easily. The Main Parts of a Digital Bank 1 - Customer Side (What users see) This includes mobile apps, websites, and ATMs. Mobile banking alone makes up 45% of all banking actions today (Statista 2024). That’s why simple design and easy navigation matter so much. 2 - Middle Layer (How things work behind the scenes) This part decides what happens when you tap a button: payments, fraud checks, and personal offers. AI chatbots here already solve 80% of customer questions without a human agent (Accenture). 3 - Back Office (Core systems) This is where accounts, transactions, and rules are managed. Digital banks like Revolut have automated 90% of back-office work, cutting costs by 40% (BCG). 4 - Partners and Integrations Digital banks connect to other companies using open APIs. For example, fintech tools like Plaid help share banking data safely. This turns banks into platforms, not closed systems. What Makes Digital Banks So Effective 1 - APIs About 85% of digital banks use open APIs for instant payments and data sharing. Traditional banks? Only 35% (Gartner). 2 - Cloud Technology Cloud systems can cut infrastructure costs by 50% and allow fast global growth. Nubank used AWS to reach 100+ million users across Latin America. 3 - AI and Machine Learning Banks use data to predict what customers need. This increases product sales per customer by 25% (McKinsey). Problems Digital Banks Still Face 1 - Security Around 60% of digital banks face cyberattacks every year (IBM). That’s why they use tools like biometrics and “zero-trust” security. 2 - Rules and Laws Regulations like GDPR and PSD2 control how data is used. Monzo spends 30% of its tech budget just on automating compliance. Old Systems Still Exist - Some banks mix new tech with old systems. - BBVA does this with its Open Platform. - Still, 70% of banks struggle with technical debt (Deloitte). What’s Coming Next Banks are becoming full ecosystems, not just places to store money. SeaBank in Indonesia combines banking with online shopping and insurance. Chime in the U.S. works with Coinbase to offer crypto services. Juniper Research says that by 2027, 60% of bank revenue will come from API-based partnerships. What do you think that are the next API services/products banks/fintechs will sell Sources: McKinsey, Gartner, BCG, IBM, Juniper Research

You can not modernise what you can not see. Most organisations have no idea what is actually running. We were brought in to uplift a monitoring estate for a critical government programme. The assumption was straightforward. Document the existing infrastructure and then rebuild it. The reality was different. Nobody knew exactly what was connected to the network. Years of mergers and upgrades had created an environment where the asset register bore little resemblance to reality. We found servers nobody remembered provisioning. We found network devices configured by contractors who left years ago. This is the discovery challenge that kills modernisation programmes. You can not upgrade systems you do not know exist. Successful discovery requires a few key things → Automated scripts that actively scan and identify every device. → Network traffic analysis to find communicating systems. → Reconciliation between documentation and reality. Our automated scripts found approximately 400 devices per day during the audit phase. Modernisation programmes that skip proper discovery build on assumptions rather than reality. What percentage of your infrastructure would discovery scripts find that is not in your asset register? #DigitalTransformation #ITInfrastructure #ShadowIT

The AI ecosystem is becoming increasingly diverse, and smart organizations are learning that the best approach isn't "open-source vs. proprietary"—it's about choosing the right tool for each specific use case. The Strategic Shift We're Witnessing: 🔹 Hybrid AI Architectures Are Winning While proprietary solutions like GPT-4, Claude, and enterprise platforms offer cutting-edge capabilities and support, open-source tools (Llama 3, Mistral, Gemma) provide transparency, customization, and cost control. The most successful implementations combine both—using proprietary APIs for complex reasoning tasks while leveraging open-source models for specialized, high-volume, or sensitive workloads. 🔹 The "Right Tool for the Job" Philosophy Notice how these open-source tools interconnect and complement existing enterprise solutions? Modern AI systems blend the best of both worlds: Vector databases (Qdrant, Weaviate) for data sovereignty, cloud APIs for advanced capabilities, and deployment frameworks (Ollama, TorchServe) for operational flexibility. 🔹 Risk Mitigation Through Diversification Smart enterprises aren't putting all their eggs in one basket. Open-source options provide vendor independence and fallback strategies, while proprietary solutions offer reliability, support, and advanced features. This dual approach reduces both technical and business risk. The Real Strategic Value: Organizations are discovering that having optionality is more valuable than any single solution. Open-source tools provide: • Cost optimization for specific use cases • Data control and compliance capabilities • Innovation experimentation without vendor constraints • Backup strategies for critical systems Meanwhile, proprietary solutions continue to excel at: • Cutting-edge performance for complex tasks • Enterprise support and reliability • Rapid deployment with minimal setup • Advanced features that take years to replicate What This Means for Your Strategy: • Technical Teams: Build expertise across both open-source and proprietary tools • Product Leaders: Map use cases to the most appropriate solution type • Executives: Think portfolio approach—not vendor lock-in OR vendor avoidance The winning organizations in 2025-2026 aren't the ones committed to a single approach. They're the ones with the most strategic flexibility in their AI toolkit. Question for the community: How are you balancing open-source and proprietary AI solutions in your organization? What criteria do you use to decide which approach fits each use case?

New research from Massachusetts Institute of Technology! The following is going to change in my opinion as more people and companies realize the advantages of open models: "Closed models dominate, with on average 80% of monthly LLM tokens using closed models despite much higher prices - on average 6x the price of open models - and only modest performance advantages. Frontier open models typically reach performance parity with frontier closed models within months, suggesting relatively fast convergence. Nevertheless, users continue to select closed models even when open alternatives are cheaper and offer superior performance. This systematic underutilization is economically significant: reallocating demand from observably dominated closed models to superior open models would reduce average prices by over 70% and, when extrapolated to the total market, generate an estimated $24.8 billion in additional consumer savings across 2025. These results suggest that closed model dominance reflects powerful drivers beyond model capabilities and price - whether switching costs, brand loyalty, or information frictions - with the economic magnitude of these hidden factors proving far larger than previously recognized, reframing open models as a largely latent, but high-potential, source of value in the AI economy."

In a merger, the word “synergy” is often used to justify the deal.  In large enterprises, that synergy usually slows down at the data layer. When two organisations combine, the Board expects a unified view of customers, margins, supply chains, and risk exposure.  What they often inherit instead is a fragmented estate: multiple Snowflake environments, parallel ERP systems, legacy SQL Servers still running critical workloads, and no shared definition of basic metrics. This fragmentation is not an IT inconvenience. It is a structural drag on EBITDA. Finance teams spend months reconciling numbers instead of integrating operations.  Procurement savings remain theoretical because spend data cannot be harmonised.  Cross-sell strategies underperform because customer records do not align.  Leadership debates whose dashboard is “correct” instead of focusing on growth. It also creates 𝐀𝐈 𝐩𝐚𝐫𝐚𝐥𝐲𝐬𝐢𝐬. Enterprises talk about Copilots, GenAI layers, and agentic automation.  But you cannot deploy intelligent workflows on top of contradictory data logic.  If “Revenue” or “Margin” means something different across business units, automation only scales inconsistency. Post-merger value realisation requires a shift from moving data to governing logic. That begins with defining a shared semantic layer before merging a single table.  1. Agree on enterprise-wide definitions.  2. Assign domain accountability.  3. Rationalise overlapping platforms.  4. Decommission legacy debt rather than stacking new cloud costs on top of old architecture. True cost synergy comes from building a disciplined, scalable data foundation that supports unified reporting, controlled cloud economics, and AI readiness. Modernization in this context is about ensuring the combined enterprise operates on one coherent data engine, so the merger becomes a multiplier of value. #MergersAndAcquisitions #DataStrategy #EnterpriseAI #DigitalTransformation #DataGovernance #BusinessStrategy

Every time I support organizations in their digital transformation path, I see the same pattern repeating. The problem is rarely about the tools. It is more often about the strategy—or the lack of it. The biggest challenges are not technological. “Lack of change management strategy,” “driving adoption,” and “culture mindset” are at the top of the list. Even “complex software” or “IT skills gaps” only become real obstacles when there is no clear vision guiding the transition. It confirms something I have been thinking for years: transformation starts from people, not from platforms. The success of digital initiatives depends on aligning leadership, mindset, and long-term planning. Without that alignment, even the best tools won't deliver impact. Let’s stop treating digital transformation like a tech upgrade and start treating it like the cultural shift it really is. #DigitalTransformation #Leadership #ChangeManagement #BusinessStrategy #Innovation #Culture #Mindset

Rewrites feel clean at the start. Then reality shows up. Missed edge cases. Broken behavior. Delayed releases. A second system nobody fully trusts. A safer option is to migrate incrementally. That’s where the Strangler Fig Pattern shines. Instead of replacing the whole legacy API at once, you put a reverse proxy in front of it and start routing traffic endpoint by endpoint. Old system keeps running. New system takes over gradually. Risk stays contained. In my example, I start with a Node.js API, add YARP as a reverse proxy, and then migrate individual endpoints into a modern .NET 10 API. The nice part is that this works just as well for old .NET Framework apps. You don’t need a giant rewrite to modernize a legacy system. You need a controlled migration path. I break down the full implementation here: https://lnkd.in/dg_zf-MV