Situation
USCIS depended on CLAIMS3, a mission-critical legacy case-management application used to manage requests for immigration benefits.
Over time, the application had accumulated substantial technical debt. Aging architecture and unreliable components made defects harder to fix, and some fixes were followed by recurring problems. The support backlog included a mix of production defects, data-quality issues, integration problems, user issues, and operational incidents.
The challenge was not simply to close more tickets. The team needed to reduce the underlying technical causes that kept creating them.
The problem
A broad rewrite would have introduced significant cost and delivery risk for a mission-critical system. At the same time, treating tickets one at a time was not changing the trajectory of the backlog.
The modernization effort needed to answer a more useful question: which parts of the system were consuming the most engineering effort and creating the most operational impact — and what architectural changes would make those areas more reliable?
Omar's role
While serving as a software architect for a federal contractor, FADLtech founder Omar Fadl provided architecture leadership and hands-on technical guidance for CLAIMS3 modernization.
For the technical-debt reduction effort, he led a small cross-functional team spanning development, testing, and business analysis. He worked with the team to identify the components creating the greatest support burden, define architectural improvements, and guide implementation.
Approach
Prioritize by impact, not by age
Rather than treating all technical debt equally, the team ranked problem areas using three practical measures:
- Developer hours consumed — how much engineering effort the area was absorbing
- Area criticality — how important the component was to the application and business process
- Ticket count — how frequently the area generated support work
Fix recurring causes, not only individual symptoms
The team re-architected unreliable services where repeated defects indicated that local fixes were no longer enough. Examples included:
- Replacing local/server-oriented biometric-image handling with a REST API interface to an enterprise repository for retrieving, editing, and saving images
- Introducing queues to decouple the application from downstream services, reducing tight dependencies and improving resilience when services were unavailable or delayed
These changes were part of a broader incremental modernization approach that improved interfaces and component boundaries without requiring a full-system rewrite.
What this demonstrates
Legacy modernization does not have to begin with a rewrite.
For systems that still provide business value, a more effective first step can be to identify the components producing the greatest operational cost, technical risk, and support burden — then modernize those areas incrementally.
That is the same principle behind FADLtech's Legacy Application Modernization Assessment: understand what is actually causing the problem, prioritize by business impact and technical risk, and create a practical path forward.
This case study describes work performed by FADLtech founder Omar Fadl while serving as a software architect for a federal contractor, prior to his current consulting work through FADLtech. Agency names are used solely to identify the context of the work and do not imply endorsement.