Modernizing Transit Systems with AWS and GenAI

Modernizing application infrastructure foundations for an Australian transport authority

Enabling secure, scalable transit operations through GenAI-assisted engineering and cloud native migrations

5 min read

Overview

A leading government transport authority operating mission-critical transit systems across a large region required highly available, secure and scalable platforms to ensure uninterrupted services. To meet these needs, it initiated the modernization program-targeted at eliminating downtime, modernizing legacy systems and driving operational efficiency without impacting existing functionality or data structures.

The Challenge

The modernization program ecosystem comprises multiple Java-based applications running on self-managed AWS EC2 infrastructure. Over time, this environment created significant friction points:

Technological obsolescence: The platform relied on an End-of-Service-Life (EOSL) technology stack—including Java 8, Spring Boot 1.x, and JUnit 4—introducing acute support and security risks.
Resilience limitations: The self-managed setup lacked modern automation, high availability (HA) and disaster recovery (DR) frameworks, leading to frequent application downtime.
Security vulnerabilities: The applications carried unresolved security bugs and structural vulnerabilities highlighted by Checkmarx analysis findings.
Compounded business impact: These constraints created unsustainable infrastructure costs, severe performance bottlenecks and manual operational overhead that extended delivery cycles and degraded commuter-facing user confidence.

The Solution

HCLTech spearheaded a comprehensive application and infrastructure modernization program, transforming the client’s software delivery lifecycle through cloud native migrations and AI-assisted engineering.

Cloud and DevOps transformation

HCLTech re-architected the legacy infrastructure, replacing high-maintenance, self-managed instances with an automated AWS-managed and serverless environment:

Modernized stack: Upgraded applications to Java 21, Spring Boot 3.2.x, Hibernate and JUnit 5.
Serverless ecosystem: Migrated core systems to AWS ECS Fargate, AWS MSK (Managed Streaming for Apache Kafka), AWS ElastiCache and AWS S3.
Infrastructure as Code (IaC): Automated deployments using CloudFormation, Ansible, Docker and YAML-based Azure Pipelines.

GenAI-led quality engineering

By embedding GitHub Copilot (GHCP) into the core engineering lifecycle, HCLTech accelerated code development, pipeline automation, IaC scripting and comprehensive unit testing. This solution was deployed via an Agile, pilot-led approach, benchmarking early velocity within a focused development group before progressively executing a phased rollout across broader transport planning and operations domains.

The Impact

The shift to a modern cloud native architecture powered by GenAI delivered substantial, quantifiable benefits across the client’s ecosystem:

High AI trust and adoption: The initiative achieved 70%-75% GenAI adoption among engineers, alongside a 60%-65% recommendation acceptance rate to standardize code delivery.
Accelerated commuter services: Engineering teams realized a 12%-18% productivity improvement, allowing the faster deployment of digital capabilities without increasing baseline capacity.
Reduced defect cycles: Quality improvements led to a 10%-15% reduction in rework, radically shortening build-to-validation timelines.
Enterprise-wide scalability: Engineers successfully achieved a 20%-30% reuse rate of AI patterns across planning, asset management and operations domains, removing duplication of effort.
Operational resilience: The elimination of legacy constraints enabled the client to reliably absorb peak transit demands and execute parallel modernization streams without risking safety or public service continuity.

Beyond the Numbers

The technical success of the modernization program highlights a powerful cultural shift for the authority's development teams. By offloading repetitive coding tasks, script generation and boilerplate infrastructure work to GitHub Copilot, the cognitive load on engineering teams has decreased dramatically. This shift has established a highly standardized execution model that dampens delivery variability across large transport initiatives. Migration away from an EOSL environment toward a serverless architecture has not only minimized operational risk but has actively restored long-term team agility, turning IT from a maintenance bottleneck into an innovation accelerator.

Celebrating Success

Client leadership highly valued the deep commitment, strong ownership of outcomes and strict accountability demonstrated by closely integrated HCLTech and client teams. The collaboration successfully hit milestones within brief windows by executing clear solution demonstrations and rapid feedback loops.

Moving forward, HCLTech plans to expand this partnership by:

Standardizing AI-driven delivery: Establishing GitHub Copilot-based modernization as the default blueprint across all customer business units.
Expanding automation use cases: Advancing GenAI and Agentic AI adoption for code analysis, defect maintenance and automated test case design.
Enabling future GTM velocity: Aligning technological roadmaps with the agency's new business initiatives to continuously enhance the overall end-customer commuter experience.