← Back to Portfolio
CargoDzen · Production

CargoDzen

A cloud-based TMS (Transportation Management System) that automates logistics operations with AI.

cargodzen.com ↗ View Architecture CI/CD Pipeline

Core Technology Stack

Python / Flask
REST API & SSE
PostgreSQL
Relational Data
Google Gemini
AI OCR & Vision
DigitalOcean
Cloud Infrastructure
Redis + Celery
Async Tasks
Docker
Containerization
Jinja2 + SSE
Real-time SSR

The Origin Story

When I began working at a logistics company in the US, I noticed how inefficient traditional workflows were. Dispatchers spent hours manually typing data from forms, and communicating load details via email or WhatsApp often led to errors and delays.

Fleet Managers faced a different struggle: trucks stayed on the road for months, creating a "blind spot" for maintenance. Drivers often forgot to report minor issues until they became major, expensive breakdowns. Existing market solutions were either prohibitively expensive or too complex for a small fleet.

I built CargoDzen to solve this: AI data entry, one communication channel, and a digital safety workflow in a single platform. I am also finalizing a native iOS application so drivers get the same tools in the cab.

System Architecture

High-level overview of the infrastructure, from load balancer to AI worker.

graph TD %% Styles classDef core fill:#1e293b,stroke:#38bdf8,stroke-width:2px; classDef service fill:#1e293b,stroke:#a855f7,stroke-width:2px; classDef default fill:#0f172a,stroke:#334155,color:#fff; User["User / Client"] -->|HTTPS| LB[Load Balancer] LB --> Web["Web Server (Flask)"]:::core %% Backend Core linked Web -->|CRUD| DB[("PostgreSQL")]:::core Web -->|Cache| Redis[("Redis")]:::core Web -->|Tasks| Celery[Celery Worker]:::core %% Services Web -->|Files| S3["DO Spaces"]:::service Celery -->|AI Request| Gemini[Google Gemini AI]:::service Celery -.->|Updates| DB

AI-Powered Order Flow

The "Smart Import" feature allows dispatchers to upload a PDF or Image of a rate confirmation. Google Gemini 2.0 Flash analyzes the document visually to extract:

  • Pickup & Delivery Addresses
  • Appointment Dates & Times
  • Load Rates ($) and Commodities
graph LR Dispatcher -->|Upload PDF| System System -->|Vision API| AI[Gemini AI] AI -->|Extract JSON| OCR OCR -->|Auto-fill| Order[Draft Order]

Key Modules

📦 Operations Hub

Centralized dashboard for dispatchers. Monitor loads from "Dispatched" to "Delivered". All documents (BOL, POD) start attached to the order.

🚛 Safety & Compliance

DOT-compliant Pre-Trip inspections. Drivers upload photos via mobile. Safety Managers get real-time alerts for defects.

👥 Role-Based Access

Granular permissions for Owners, Dispatchers, Drivers, and Accountants. Invite team members via email.

💬 Secure Messenger

Internal encrypted chat replacing WhatsApp. Isolates drivers from each other while keeping office comms open.

💰 Finance & Analytics

Real-time revenue dashboards, driver settlement calculations (Gross/Miles), and batch document downloads for factoring.

Safety Inspection Workflow

graph TD Driver -->|Daily Form| App App -->|Upload Photos| S3["DO Spaces"] App -->|Defects?| Check{"Issues?"} Check -->|No| OK[Status: OK] Check -->|Yes| Alert[Alert Safety Mgr] Alert -->|Repair| Mechanic Mechanic -->|Fix| Resolve[Resolved]

DevOps & Security

🛡️ Audit & Compliance

Full JSON audit logs of every action (Time, IP, User-Agent) for security investigations and enterprise compliance.

🏗️ Zero Downtime Architecture

Blue/Green deployment strategy ensures users never experience downtime during updates. Health checks automatically route traffic to healthy instances.

CI/CD Pipeline Workflow

Jenkins runs the delivery pipeline from commit to production, with tests gating every build.

graph TD %% Define Styles classDef ci fill:#1e293b,stroke:#38bdf8,stroke-width:2px; classDef cd fill:#1e293b,stroke:#a855f7,stroke-width:2px; classDef default fill:#0f172a,stroke:#334155,color:#fff; Dev[Developer] -->|Push| Git[Git / GitHub] Git -->|Webhook| Jenkins[Jenkins Server] %% CI Phase Jenkins --> Source[Checkout Code]:::ci Source --> TestEnv[Deploy to Test]:::ci TestEnv --> PyTest[Run Auto Tests]:::ci %% CD Phase PyTest -->|Success| Docker[Build Image]:::cd Docker -->|Push| Registry[Container Registry]:::cd Registry -->|Deploy| Prod[Production]:::cd Prod -->|Verify| Health[Health Checks]