Technical Architecture

Gemini 3 Flash with
agentic function calling

3-tier reconciliation with dynamic thinking levels, 8 investigation tools, streaming reasoning via Firestore, and pattern memory with Levenshtein fuzzy matching.

Start Free Trial View Pricing

Full stack

System Architecture

Next.js 14 frontend, Firebase Cloud Functions backend, Gemini 3 Flash AI, and Firestore for real-time streaming.

Client

Next.js 14 + React 19 + Firestore SDK + Tailwind CSS

Real-time onSnapshot listeners for streaming AI progress

Real-time listeners

Firebase Cloud Functions (onCall)

Vision Pipeline

PDF scan (inlineData)
CSV detection & parsing
Template fingerprinting
responseMimeType: JSON

Reconciliation Engine

Tier 1: Rule-based scan
Tier 2: thinking_level LOW
Tier 3: thinking_level HIGH
Pattern memory + learning

AI Assistant

queryFirestore tool
getOverview / getCashFlow
90-day cash flow forecast
Natural language queries

Firestore

transactions, invoices, bills, vendor_patterns, match_history, fx_cache, reconciliation_runs

ProgressStream

Firestore events → onSnapshot → Terminal UI with color-coded reasoning steps

Gemini 3 API

How We Use Gemini 3

Four distinct Gemini 3 API features — Vision, Structured Output, Function Calling, and Thinking Levels — each serving a specific role.

Vision Processing

PDF pages sent as base64 to Gemini 3 Flash's multimodal input. Detects tables, headers, and transaction rows from any bank format.

inlineData: {
  mimeType: "application/pdf",
  data: base64EncodedPDF
}

Structured Output

Guaranteed JSON responses — no regex parsing of AI text. Every extraction and reconciliation returns typed, validated data.

config: {
  responseMimeType: "application/json",
  // Gemini returns valid JSON directly
}

Function Calling

8 tool declarations registered with Gemini. Manual calling loop processes tool calls until the AI is satisfied — up to 10 iterations.

tools: [{
  functionDeclarations: [
    searchTransactions,
    findInvoiceCombination,
    getFXRate, // ... 8 tools total
  ]
}]

Thinking Levels

Dynamic thinking budget per tier. LOW for batch matching (speed). HIGH for complex investigations (depth). No thinking for rule-based.

thinkingConfig: {
  thinkingLevel: ThinkingLevel.HIGH
  // LOW for Tier 2, HIGH for Tier 3
}

Deep dive

Reconciliation Engine

Five scoring signals, fee detection, FX conversion, combined payments, and pattern memory — all in one engine.

5-Signal Scoring Formula

Reference Match/40

Invoice # found in description

Amount Match/35

Exact, FX-converted, or fee-adjusted

Identity Match/25

Vendor name fuzzy match (Levenshtein)

Time Proximity/20

Days between invoice and payment

Context/5

Unique amount, cross-currency bonus

Total: /125 → normalized to 0-100% confidence

Fee Pattern Detection

Stripe2.9% + $0.30

PayPal2.9% + $0.30

Square2.6% + $0.10

Wise1%

Card3%

When a transaction amount doesn't match exactly, the engine checks if it matches after subtracting known processor fees.

Multi-Currency FX

40+ currency pairs with 3-tier fallback: Live API → Firestore cache → hardcoded rates
Pegged currency detection — ANG, AWG, XCG treated as equivalent
Cross-currency via USD pivot when direct rate unavailable
FX tolerance: 2% for rounding, 5% for rate variance, 10% for fluctuation

Combined Payment Detection

When one transaction pays multiple invoices, the engine runs subset-sum with early termination to find document combinations that match the payment amount.

Max 5 documents per combination (configurable)
Groups by counterparty first (most likely), then cross-vendor
Early termination after 3 matches found
2% tolerance for rounding differences

// Subset-sum with early termination
function findCombination(
  docs, targetAmount, tolerance,
  onFound, maxDocs = 5
) {
  const recurse = (startIdx, current, sum) => {
    if (found >= 3) return; // early termination
    if (sum > target + tol) return; // prune
    if (current.length >= 2 && diff <= tol)
      onFound(current, sum);
  };
}

Specifications

Technical Specifications

AI Model	Gemini 3 Flash
Thinking Levels	LOW (batch) / HIGH (investigation)
Agent Tools	8 function-calling tools
Max Iterations	10 per investigation
Batch Size	10 txns, 4 concurrent batches
FX Currencies	40+ pairs, 3-tier fallback
Fee Patterns	Stripe, PayPal, Square, Wise, Card
Fuzzy Matching	Levenshtein distance, 0.7 threshold
Auto-Confirm	93%+ confidence
Streaming	Real-time Firestore ProgressStream
Time Budget	4 min/batch with graceful degradation
Combined Payments	Subset-sum, max 5 docs, early termination
Cash Flow Forecast	90-day, 3 periods, 4096-token thinking budget
Anomaly Detection	2σ outliers, slow payers, FX exposure, match rate
Chat Agent	Dynamic queryFirestore + 11 collections + 10 rounds
Chat Thinking	2048-token budget, 90s timeout, 20-msg history

Intelligence layer

Anomaly Detection & Cash Flow Forecast

Beyond reconciliation — statistical analysis, risk scoring, and predictive forecasting powered by Gemini 3 extended thinking.

Anomaly Detection

Statistical Outliers (2σ)

Debit amounts exceeding mean + 2 standard deviations flagged automatically. Runs on every transaction set.

Slow Payer Identification

Vendors with 30+ day delays detected from AI-learned vendor_patterns. Surfaces cash flow risks.

Low Match Rate Alerts

>20% unreconciled transactions triggers high-priority alert. Uses latest reconciliation_run data.

FX Exposure Detection

Cross-currency matches flagged for rate variance review. Detects when txn.currency ≠ doc.currency.

Spending Trend Analysis

Monthly debit comparison: >10% increase = trending up, >10% decrease = trending down.

// insights/page.tsx — computed on real data
const anomalies: { type, severity, description }[] = [];
// Statistical outlier detection
if (amount > mean + 2 * stdDev) anomalies.push(...);
// Slow payer from vendor_patterns
if (p.typicalDelay.max > 30) anomalies.push(...);

90-Day Cash Flow Forecast

Data Collection (6 parallel queries)

accounts

transactions (100)

invoices

bills

vendor_patterns

reconciliation_runs (3)

Gemini 3 Flash Configuration

Modelgemini-3-flash

Temperature0.3 (conservative)

Thinking Budget4,096 tokens

Response Formatapplication/json

Max Output4,096 tokens

Output Structure

3 forecast periods — 30/60/90 days with confidence scores

Risk factors — severity + estimated dollar impact

AI recommendations — actionable next steps

Reasoning — 2-3 paragraph methodology explanation

Thinking Animation (7 steps)

✓ Analyzing historical transaction patterns...

✓ Scanning upcoming receivables and payables...

✓ Detecting seasonal spending trends...

✓ Computing projected inflows from unpaid invoices...

✓ Estimating recurring expense patterns...

✓ Assessing cash flow risk factors...

● Generating 90-day forecast with Gemini 3...

Agentic assistant

Financial Chat — “Ask Your Books”

An autonomous AI agent with direct Firestore access. Gemini 3 constructs its own queries from a full database schema — no hardcoded search functions.

Dynamic Query Architecture

queryFirestore — The Primary Tool

AI constructs arbitrary queries with filters, ordering, text search, and field selection. userId filter auto-applied for security.

collectionfilters[]orderByorderDirectionlimitselectFields[]textSearch

Security Whitelist — 11 Collections

invoicesbillstransactionsaccountsstatementsreconciliation_matchesvendor_patternsinvoice_line_itemsreconciliation_runsmatch_historydocuments

Agent Loop

Max iterations10 rounds

Thinking budget2,048 tokens

Conversation historyLast 20 messages

Timeout90 seconds

Real-Time Progress Streaming

chat_progress — live events

> ANALYZING QUERY

Processing: "Which vendors have unpaid invoices?"

> QUERYING DATABASE

Querying invoices: paymentStatus == unpaid

Found 12 results in invoices collection

Querying vendor_patterns: orderBy matchCount desc

Loaded financial overview across all collections

AI analyzing results (round 2)...

> GENERATING RESPONSE

Complete — 3 queries, 2 rounds, 2.1s

financial-chat.ts — agentic loop

// Manual function-calling loop
while (response.functionCalls?.length > 0
  && iterations < 10) {
  for (const fc of response.functionCalls) {
    await writeProgressEvent(...);
    result = await executeTool(fc.name, fc.args);
  }
  response = await ai.models.generateContent(...);
}

Export to any accounting system

Once your data is extracted, export it in the exact format your software needs.

Standard Formats

Ready-to-use templates for common accounting software.

Excel (.xlsx)CSVJSONPDF Report

Accounting Software

Pre-configured formats that import directly.

QuickBooksXeroSageFreshBooks

Custom Templates

Create your own export format with custom columns.

Custom CSVAPI/JSONBank Format

Enterprise Security

SOC 2 Type II

Certified compliant

256-bit Encryption

At rest and in transit

99.9% Uptime

SLA guaranteed

GDPR Compliant

Full data control

Ready to see it work?

Start your free 14-day trial today. No credit card required.

Start Free Trial Request Demo

// Subset-sum with early termination function findCombination( docs, targetAmount, tolerance, onFound, maxDocs = 5 ) { const recurse = (startIdx, current, sum) => { if (found >= 3) return; // early termination if (sum > target + tol) return; // prune if (current.length >= 2 && diff <= tol) onFound(current, sum); }; }

// insights/page.tsx — computed on real data const anomalies: { type, severity, description }[] = []; // Statistical outlier detection if (amount > mean + 2 * stdDev) anomalies.push(...); // Slow payer from vendor_patterns if (p.typicalDelay.max > 30) anomalies.push(...);

// Manual function-calling loop while (response.functionCalls?.length > 0 && iterations < 10) { for (const fc of response.functionCalls) { await writeProgressEvent(...); result = await executeTool(fc.name, fc.args); } response = await ai.models.generateContent(...); }

Gemini 3 Flash withagentic function calling

System Architecture

How We Use Gemini 3

Vision Processing

Structured Output

Function Calling

Thinking Levels

Reconciliation Engine

5-Signal Scoring Formula

Fee Pattern Detection

Multi-Currency FX

Combined Payment Detection

Technical Specifications

Anomaly Detection & Cash Flow Forecast

Anomaly Detection

90-Day Cash Flow Forecast

Financial Chat — “Ask Your Books”

Dynamic Query Architecture

Real-Time Progress Streaming

Export to any accounting system

Standard Formats

Accounting Software

Custom Templates

Enterprise Security

SOC 2 Type II

256-bit Encryption

99.9% Uptime

GDPR Compliant

Ready to see it work?

Gemini 3 Flash withagentic function calling

System Architecture

How We Use Gemini 3

Vision Processing

Structured Output

Function Calling

Thinking Levels

Reconciliation Engine

5-Signal Scoring Formula

Fee Pattern Detection

Multi-Currency FX

Combined Payment Detection

Technical Specifications

Anomaly Detection & Cash Flow Forecast

Anomaly Detection

90-Day Cash Flow Forecast

Financial Chat — “Ask Your Books”

Dynamic Query Architecture

Real-Time Progress Streaming

Export to any accounting system

Standard Formats

Accounting Software

Custom Templates

Enterprise Security

SOC 2 Type II

256-bit Encryption

99.9% Uptime

GDPR Compliant

Ready to see it work?

Gemini 3 Flash with
agentic function calling

Gemini 3 Flash with
agentic function calling