The Problem
Argentina's economic context is highly volatile. Financial information is scattered and inconsistent: multiple dollar exchange rates, inflation, country risk, and key indicators appear fragmented, with unclear definitions and difficult comparisons.
People making financial decisions—from small savers to businesses—lack a single source of truth.
My Role
Founder & Solo Developer — Built the entire product from zero:
- Vision & Strategy — Defined the product direction and market positioning
- Problem Research — Deep-dived into user pain points around financial data
- Roadmap & Prioritization — Decided what to build and when
- Full-Stack Development — Designed and built the entire platform end-to-end
- Data Engineering — Built aggregation system with 5+ sources
- User Experience — Transformed complex data into clear, actionable information
The Solution
A data-driven fintech platform that consolidates Argentina's key financial indicators:
- Multiple Dollar Rates — Blue, MEP, CCL, official, crypto, and more in one view
- Real-time Updates — 30-second refresh during market hours
- Data Reliability — Cross-references 5+ sources with confidence indicators
- Historical Context — Trends and comparisons over time
- Actionable Insights — Not just numbers, but what they mean
Technical Architecture
| Layer | Technology |
|---|---|
| Framework | Next.js 15 (App Router) |
| Language | TypeScript (strict mode) |
| Database | Neon Postgres |
| State Management | TanStack Query |
| Auth | Custom JWT + CSRF protection |
| Updates | 30-second intervals |
| Testing | 474 E2E tests |
Scale & Complexity
| Metric | Count |
|---|---|
| API Routes | 50+ |
| Custom Hooks | 80+ |
| E2E Tests | 474 |
| Data Sources | 5+ |
Data Strategy
DB-first, not API-first. Instead of trusting any single source:
- Aggregate multiple sources in real-time
- Detect anomalies and flag deviations automatically
- Weight sources by historical reliability scores
- Show uncertainty transparently with confidence indicators
- Cached fallback with decay indicators when sources fail
Authentication Architecture
Custom-built auth system prioritizing security:
- JWT tokens with short expiration
- CSRF protection on all mutations
- Session management with secure cookie handling
- No third-party auth dependencies
Key Learnings
"More sources ≠ better accuracy"
More sources meant more edge cases. Some APIs update at different times. Some have bugs during weekends. The "consensus" was often wrong.
Solution: Quality over quantity. Weight sources by historical reliability.
"Users want trust, not precision"
A value with "updated 2 minutes ago, high confidence" is more useful than a "precise" number with unknown freshness.
Solution: Built confidence displays from day one. Transparency beats false precision.
"474 tests aren't overkill for fintech"
Financial data requires absolute reliability. Users make real money decisions based on this data.
Solution: Comprehensive E2E test coverage for every data flow and edge case.
Iteration Philosophy
The platform is alive—constantly iterated based on real usage and user feedback. Data-driven decisions about the product that provides data-driven decisions.
What I'd Change
- Start with fewer, higher-quality sources
- Add user-reported corrections as a signal earlier
- Build community features for power users