Fast-Track Implementation Guide

Setup Guide

Deploy the WebApp, pair your Android device via CameraX QR, configure local GGUF models, and unlock automated financial intelligence in 4 easy steps.

1

Deploy & Register Web Dashboard

Launch the platform on your infrastructure (Docker Compose or cloud host such as Railway) and register your administrative account.

  1. If running self-hosted, clone repository and configure your .env database credentials.
  2. Run docker compose up -d to spin up WebApp (port 8080), MySQL (port 3306), and ML Service (port 9050).
  3. Navigate to the Registration Page and create your master account.
  4. Log into the dashboard to access the Control Center.
Tip: The web dashboard includes automatic schema migration on startup.
2

Install Companion App & Pair via QR Code

Install the Android companion APK and pair it seamlessly with your account using the vertical CameraX barcode scanner.

  1. Download and install the APK on your Android device (Android 8.0+ supported).
  2. In the WebApp dashboard, navigate to Control Center → Devices.
  3. Click Pair New Device to generate a unique pairing QR code with your encrypted server token.
  4. Open the Android app, tap Scan Pairing QR, and align the camera view with your screen.
  5. The app automatically saves the server URL, performs cryptographic handshake, and stores the token securely.
Security: The server stores only the SHA-256 hash of your device token. You can revoke any device instantly from the dashboard.
3

Activate Local GGUF Model Preset

Enable local semantic reasoning via the ML management console without needing external LLM APIs.

  1. In the WebApp sidebar, go to Admin → ML Management → Models.
  2. Review available model presets (e.g. Qwen 2.5 3B Instruct Q4_K_M).
  3. Click Download / Activate — the system verifies disk space and memory headroom before initializing.
  4. Verify model status in Admin → Telemetry to confirm llama-server process health and RSS memory usage.
Performance: The ML microservice uses AVX2 CPU vectorization automatically, running efficiently even on standard 4-core VPS instances.
4

Sync SMS & Run Semantic Analysis

Initiate your initial SMS batch upload and generate your structured financial ledger.

  1. On the Android app, tap Sync SMS History to queue existing mobile money messages into local SQLite.
  2. WorkManager batches and AES-256 encrypts the payload, transmitting it to /api/v1/upload.
  3. In the WebApp dashboard, click Rescan / Analyze to trigger the semantic classification cycle.
  4. Monitor the live modal progress dialog as messages are categorized into sent, received, utilities, airtime, and Fuliza.
  5. Explore your interactive spending charts, financial health score, and exportable ledgers.
\n