Back to Home
Project 01

MaxiCLI

WhatsApp commerce platform · AI-powered

A full platform that turns WhatsApp into an AI-powered sales and customer-service channel — dashboard, automatic orders, reports, and an AI assistant for your team.

Role
Solo architect & lead engineer
Status
Built to production-grade
Period
2024 — 2025
Domain
Conversational commerce
FastAPINext.js 15PostgreSQLpgvectorRedisWebSocketRAGMulti-tenant
The Problem

Small businesses running on WhatsApp lose orders to the chaos: messages scattered across phones, no record of what was promised, no numbers to look at, and everything done by hand. Ready-made chatbots are too rigid, and building your own takes forever.

Approach

One platform where every business's data stays fully separate, an AI assistant that proposes actions but never runs them on its own, and a foundation built to keep growing without breaking.

Architecture

Tenancy & data

PostgreSQL with row-level scoping per tenant, pgvector for retrieval, Redis for queues and ephemeral state. Alembic migrations tied to spec docs.

Agent layer

Dual-model setup — a fast model for routing and extraction, a stronger model for reasoning. Tool-calling with idempotency keys and structured pending actions.

Realtime

WebSocket fan-out to the admin dashboard, with optimistic updates and re-sync on reconnect.

Quality

220+ test files covering domain rules, schema constraints, and agent contracts before they reach production.

Outcomes
  • Built to production quality end to end — ready to serve real businesses.
  • Work that used to take hours became a single confirmation click.
  • Structured so new features ship fast without breaking what already works.
Stack
FastAPINext.js 15PostgreSQLpgvectorRedisRAGWebSocketsAlembicDocker