We will be undergoing planned maintenance on January 16th, 2026 at 1:00pm UTC. Please make sure to save your work.

Credexia

turns loan documents into instant , trusted compliance decisions

Comment

Credexia - THE OPERATING SYSTEM FOR LOANS

INSPIRATION

We analyzed the Loan Market Association ecosystem — tools like LMA Automate for drafting and trading platforms like Octaura for price discovery — and identified a massive white space in between.

While banks can agree on trades and sign loans digitally, post-closing servicing and settlement are still stuck in the Stone Age:

  • Fragmented Excel sheets
  • Endless email chains
  • Manual data re-entry
  • Broken reconciliation

Liquidity is not blocked by lack of capital. It is blocked by operational friction.

Our goal: Build the Operating System for Loans — a unified platform connecting covenant monitoring, trade settlement, and cash distribution in real time.

WHAT Credexia DOES

Credexia is a unified desktop platform covering three critical loan lifecycle stages in one dashboard.

Credexia - MONITORING

  • Tracks Financial covenants (Debt / EBITDA)
  • Tracks ESG covenants (Carbon Emissions)
  • Users upload a PDF financial report
  • AI extracts values and calculates compliance

Traffic Light System:

Green = Safe Amber = Warning Red = Breach

Auto-Pricing: If an ESG covenant is breached, the interest rate is automatically adjusted according to loan terms.

TRADECLEAR - SETTLEMENT

  • Digitizes transfer of loan ownership
  • Users upload a Notice of Assignment PDF
  • AI extracts Seller, Buyer, and Trade Amount
  • After approval, the Ownership Register updates instantly
  • Removes manual ledger updates entirely

AGENTFLOW - CASH DISTRIBUTION

  • Automates the Payment Waterfall
  • Distributes borrower payments across lender banks
  • Enforces zero rounding error
  • Any remainder goes to the Agent Bank
  • Generates SWIFT-ready CSV in seconds

GLOBAL STATE

All three modules share a Global State.

If ownership changes in TradeClear, AgentFlow instantly reflects the new owner without a page reload.

HOW WE BUILT IT

Frontend: Next.js (App Router) Backend: Supabase (Database + Storage)

AI Layer:

  • Gemini API with 4-tier fallback: Gemini 2.5 Flash Gemini 2.5 Pro Gemini 1.5 Flash Gemini 1.5 Pro
  • OCR.space as raw text backup
  • Confidence Gate: if below 60%, user must confirm

Logic Layer:

  • All business logic implemented as pure functions
  • Deterministic outputs
  • Payment Waterfall guarantees Sum(Shares) = Total Cash

Blockchain Layer:

  • Polygon Amoy Testnet
  • SHA-256 covenant sealing
  • Demo Mode using FNV-1a hash simulation when network fails

State Management:

  • In-memory global singleton: global.__tradeState
  • Prevents loss during Next.js hot reload

CHALLENGES

Private Key Fiasco: .env key had an extra character. Ethereum keys must be exactly 64 hex characters.

OCR Reality: Messy financial tables broke OCR. Solution: Demo-friendly PDFs + confidence gating.

Schema Drift: Frontend expected fields that did not exist in Supabase. Frontend adapted to database instead of migrating DB mid-hackathon.

Floating Point Errors: Solved by rounding strictly at final step and assigning remainder to Agent Bank.

Model Name Confusion: Incorrect Gemini model names caused 404 errors. Solved with retry fallback system.

WHO THIS IS FOR (User Base)

Credexia serves:

  • Credit Risk Analysts
  • Loan Agency Teams
  • Syndicated Loan Traders
  • Treasury Operations
  • ESG Compliance Teams
  • Audit & Regulatory Teams

And eventually:

  • Rating agencies
  • Regulators
  • Institutional investors

ACCOMPLISHMENTS

Zero-Variance Payment Waterfall Engine: Handles fractional splits and guarantees perfect totals.

Traffic Light UI: Instant risk visibility across 50+ loans.

Honest Error Handling: No silent failures. Users see real confidence and validation errors.

Resilient Architecture: App never crashes during demo regardless of API failure.

Unified State: Trade settlement instantly updates cash distribution.

BUSINESS MODEL (Monetization)

Credexia can monetize via:

  • Per-loan subscription
  • Per-document processing
  • Per-settlement fee
  • Compliance enterprise licenses
  • ESG pricing modules
  • Audit trail services

This is not a consumer SaaS. This is institutional infrastructure.

WHAT WE LEARNED

  1. The last mile in fintech is process, not math.
  2. AI needs guardrails.
  3. Simulation beats broken reality.
  4. Finance demands deterministic precision.

END

Built With

Share this project:

Updates