Frontend
React TypeScript TanStack Query
01 — FinTech SaaS
PaySync
B2B payment reconciliation platform automating multi-currency matching for finance teams at Series A+ startups.
product screenshot — dashboard overview
The Challenge
Manual reconciliation was eating 30+ hours a week.
Loopcast's finance team was manually matching thousands of multi-currency transactions across five payment providers every month. Spreadsheet-based workflows couldn't keep pace with their growth, and errors were slipping into investor reporting.
They needed a system that could ingest data from disparate sources, apply fuzzy matching logic across currencies and timing gaps, and surface exceptions for human review — without becoming another tool nobody trusted.
The Solution
An automated matching engine with human-in-the-loop review.
We built a rules-based matching engine that ingests transactions via the Stripe API and bank feed integrations, normalizes currencies against daily FX rates, and auto-resolves ~90% of transactions with zero human input.
A review queue surfaces the remaining exceptions with full audit trails, and a PostgreSQL-backed reporting layer gives finance leadership real-time visibility into reconciliation status across every provider.
Product gallery
reconciliation queue — screenshot
exception detail — screenshot
reporting dashboard
mobile view
audit trail
Results
62%
Reduction in manual reconciliation time
90%
Transactions auto-resolved with no human review
14 wks
From kickoff to production launch
Tech stack
Backend
Node.js Express BullMQ
Data
PostgreSQL Redis Stripe API
Infra
AWS ECS Docker GitHub Actions
CrackinCode shipped our MVP in 6 weeks. We went from a Figma prototype to a live product that closed our pre-seed round.
Sarah ChenCo-founder & CEO, Loopcast · San Francisco, USA