The problem
A growing fintech was bottlenecked by a homegrown internal tool. Customers had no self-serve, ops was drowning in tickets, and every new feature added weeks of manual work.
The approach
- 01
Two-week discovery
User journeys, data model, compliance constraints, and a clickable prototype the team could test against. Scope cuts agreed before any production code.
- 02
Edge-first stack
TanStack Start, Postgres with row-level security, and a typed RPC layer end-to-end. Production-grade defaults out of the box.
- 03
Customer + admin shipped together
Role-based access, billing portal, audit log, and an admin console — shipped together so ops and customers benefit on day one.
- 04
Observability from day one
Structured logs, SLO dashboards, and a quiet on-call rotation. Production isn't a victory if you can't see what it's doing.
Results
Tickets (relative)
57% baseline
Active users
12400
Uptime
99.99%
The outcome
Eight weeks from kick-off to a paying-customer-facing portal. Support tickets down 43% in the first quarter; 99.99% uptime across the first six months.
"We've worked with three agencies. Contingo is the first one that shipped on time, on scope, and left us with code we actually want to keep."