I'm Elhoucine Maskour — Chief Technology Officer at Africodex (Marrakech, Morocco) and a hands-on full-stack engineer with 6+ years across system architecture and technology leadership.
I architect and ship production platforms — high-throughput APIs, real-time data pipelines, multi-tenant SaaS, and agentic AI systems — across trading, ERP, and marketplace domains. Today I direct the engineering organization, the cloud practice, and the Microsoft CSP portfolio behind them.
- 🔭 Currently building an Agentic Business Automation Platform — a multi-agent system that turns plain-language business requests into executed actions across Odoo, email, WhatsApp, and Meta
- 🏗️ Leading 15+ engagements for 12+ enterprise clients across Morocco and Africa
- 🌱 Deepening my work in LangGraph multi-agent orchestration, RAG pipelines, and Rust (Actix, Rocket)
- 💬 Ask me about FastAPI, NestJS, LangChain/LangGraph, Odoo, PostgreSQL, and Docker
- 🎓 MSc Data Science · Ibn Zohr University
- 📫 Reach me at elhoucine.maskour7@gmail.com
- 🗣️ Arabic (native) · English (professional) · French (professional)
Languages
Backend & Frameworks
ML & AI
Frontend
Databases & Search
Async & Messaging
DevOps & Cloud
ERP & Cloud Administration
🤖 Agentic Business Automation Platform — Africodex SaaS product
Multi-agent AI system that turns plain-language business requests into executed actions across Odoo, email, WhatsApp, and Meta. Built a LangGraph supervisor that plans, sequences, and summarizes specialist agents across RAG, Odoo integrations, marketing content, communications, analytics, and social publishing — with a dedicated Odoo pipeline handling record reads and writes through entity resolution, confirmation gates, and multi-step workflows.
FastAPI · LangGraph · React 19 · PostgreSQL · Redis · ChromaDB · MinIO · Meta API
🏢 Morocom — Africodex SaaS product
Multi-tenant SaaS ERP running day-to-day commercial operations for 8 client businesses. Architected a pnpm monorepo around a tenant-scoped NestJS 10 API serving four React 18 front-ends — back office, customer portal, delivery portal, and tenant dashboard — covering inventory, partners, orders, quotes, invoicing, payments, deliveries, and financial reporting.
NestJS 10 · TypeORM · PostgreSQL · React 18 · pnpm monorepo
💬 Marnova — client: ProTrain, professional training company (Marrakech)
Outbound campaign and two-way messaging platform built on Meta's official WhatsApp Business Cloud API. Developed a FastAPI backend for authentication, Excel contact ingestion, campaign scheduling, delivery, and inbound webhooks via Celery and Redis — with a React 18 PWA for analytics and an OpenAI-powered assistant that auto-answers incoming messages.
FastAPI · SQLAlchemy/MySQL · Celery · Redis · React 18 · Vite · Tailwind · Docker Compose
🏘️ Terramez — client: Global Gestion (Marrakech)
Enterprise real-estate marketplace built for the Moroccan property market. Delivered a Dockerized Turborepo monorepo where one NestJS API serves a Next.js 14 SSR/ISR portal, a React admin back office, and a parallel construction (BTP) office — covering listing publication, moderation, sub-100 ms faceted search, and Stripe-paid visibility packs under JWT role-based access control.
NestJS · PostgreSQL · TypeORM · Redis · BullMQ · Meilisearch · MinIO · Next.js 14 · Stripe
| Role | Company | Period |
|---|---|---|
| Chief Technology Officer | Africodex · Marrakech | May 2026 – Present |
| Full-Stack Engineer & Data Scientist | Dataikyo · Casablanca | Oct 2022 – May 2026 |
| Full-Stack Engineer | ABA Technology · Casablanca | Sep 2021 – Sep 2022 |
| Full-Stack Developer | YAM Tech · Marrakech | Oct 2020 – Sep 2021 |
Highlights
- Own the technology organization end to end, leading 5+ engineers across architecture, delivery, and engineering standards
- Deliver tailored Odoo ERP implementations — module engineering, data migration, and integration with incumbent business systems
- Drive Microsoft CSP engagements as an authorized partner: Microsoft 365, Windows Server, Azure infrastructure, and data center migrations
- Engineered multi-layer caching strategies that lifted data retrieval efficiency by up to 40%
- Automated invoice and receipt processing via NLP document extraction — 30% less manual intervention, 25% higher data accuracy
- Refactored database queries and indexing strategy, cutting database load 30% and improving response times 50%
| MSc, Data Science · Ibn Zohr University, Agadir | 2018 – 2020 |
| BSc, Computer Science · Ibn Zohr University, Ouarzazate | 2014 – 2018 |
| IBM Generative AI Engineering · Coursera | 2025 |
| Deep Learning Specialization · Coursera | 2021 |
| Full-Stack Engineer · Coursera | 2020 |
