Malaysian IT teams adopt incremental AI modernization to cut downtime risk
Enterprise IT teams in Malaysia are shifting from one-off cloud migrations to continuous, modular upgrades to keep pace with AI demands, industry executives say.
Source: RSS · July 31, 2026 at 3:02 AM · AI-assisted report
KUALA LUMPUR, 31 JULY 2026 —
Listen to this article
DomainFork Audio · read aloud
Enterprise IT teams in Malaysia are shifting from one-off cloud migrations to continuous, modular upgrades to keep pace with AI demands, industry executives say.
According to a recent McKinsey study cited in trade media, 88% of global organizations now embed AI in at least one function, but legacy systems built for stability struggle with real-time, data-heavy workloads. Cognizant’s senior vice-president Hari Parameswaran told ETCIO that Malaysian delivery units are using the Strangler Fig pattern—layering modern microservices alongside core systems—to avoid the outages that cripple revenue in always-on markets.
“Large-scale rip-and-replace programmes risk operational downtime and customer disruption,” Parameswaran said. “An incremental, business-outcome-led approach lets teams modernise component by component while the business keeps running.” He added that Malaysian clients typically prioritise customer-facing features first, then back-end data pipelines, aligning spend with immediate revenue impact rather than long-term architecture aesthetics.
Modernisation hinges on three pillars, he said: data liquidity, modular interfaces, and cloud-native resilience. Legacy silos often trap data in proprietary formats; Parameswaran warned that without seamless integration, AI models cannot ingest clean, timely signals. “AI workloads need elastic infrastructure—containers, serverless, and APIs—to scale on demand,” he added.
Generative AI is now accelerating the process itself. Teams in Malaysia use AI tools to auto-document undocumented COBOL routines, refactor monoliths, and generate synthetic test data, creating a feedback loop where smarter code enables smarter AI. “We’ve seen projects cut manual effort by 40%, letting engineers focus on higher-value work,” Parameswaran said.
The long-term discipline matters more than the technology stack, he argued. “Future-proofing must become an operating standard—rejecting proprietary shortcuts, insisting on open interfaces, and moving from break-fix IT to predictive operations.” Malaysian enterprises embedding these principles now treat application modernisation as a continuous capability, not a one-time migration.
Malaysia Impact
Global development — watch for knock-on effects on oil prices, the ringgit, and KLCI risk sentiment.