Project 1 — Payments
Scope: Payments • Reliability • Ops & Support
Tools: SQL, Amplitude/GA, Jira, Figma
Payout Reliability: Fewer Failures, Faster Resolution
Problem: Transfers were failing too often and users didn’t understand why. Support volume was rising and provider performance was inconsistent.
What I did:
Mapped failure modes (validation, timeouts, provider errors, reversals) and defined a success-rate dashboard
Improved pre-transfer validation + error messaging to reduce avoidable failures
Introduced retry rules and smarter provider routing based on performance
Added operational alerts for “stuck” transfers and clearer status updates for users
Impact:
28% reduction in failed transfers
18% fewer payment-related support tickets
Project 2 — Risk
Scope: Risk • Fraud • Compliance-by-design
Tools: SQL, Looker/Data Studio, Jira, Figma
Risk-Based Verification: Safer Transactions Without Added Friction
Problem: Fraud losses were creeping up, but adding blanket friction would hurt conversion and trust.
What I did:
Partnered with risk/compliance to define event triggers and acceptable controls
Designed step-up verification only for high-risk moments (new device, unusual amount, suspicious velocity)
Built a simple rules framework + review workflow for ops
Instrumented funnel drop-off and loss metrics to tune thresholds post-launch
Impact:
20% reduction in fraud loss rate
Conversion held steady through targeted step-ups
Improved auditability and incident review