Cellulant Kenya Limited · Senior QA Engineer in Test · Jan 2025 – present
The customer problem
The same classes of issues kept reaching customers. Support would resolve a ticket, the customer would move on — and weeks later a different customer would hit the same failure through a different door. Each repeat incident cost support time and eroded customer confidence in the platform.
What I investigated
I acted as the bridge between support and engineering: tracing customer-reported incidents across logs, monitoring tools, API responses, and deployment records to find the real origin. Every issue got mapped to a root-cause category, so patterns became visible instead of anecdotal.
What I built and shipped
- A systematic pipeline converting customer-reported issues into preventive regression coverage — automated where possible, manual where it mattered.
- Configurable automated testing workflows inside CI/CD, so fixes arrive with proof they hold.
- A stakeholder-facing dashboard integrating ClickUp and TestRail data, giving non-technical stakeholders early visibility into quality trends and account-level risk.
Outcome: recurring production defects down 20% within six months; wider adoption of automated testing across core services; faster, calmer escalation cycles because repeat issues stopped recurring.
What I’d tell a customer success team
Defect recurrence is churn pressure with a delay on it. The work that reduces it — root-cause discipline, prevention pipelines, visible health metrics — is retention work, whatever job title it hides under.