Mouhcine Toumi is an AI & Data Engineer who turns AI ambitions into production systems that give a business real leverage.
His work runs in three lanes: AI over a company's own data (retrieval-augmented generation, agents, and extraction from messy documents); data pipelines and integrations (ingestion, modeling, and the quality checks that make the numbers trustworthy); and full product builds, from first prototype to the version customers depend on.
Recent projects include a content pipeline that turns one draft into a scheduled week of platform-native posts, a document-RAG backend that returns cited answers from a pile of PDFs, and a multi-tenant gym platform shipped to production behind CI/CD. He takes full ownership of what he builds, is candid about trade-offs, and cares most about the invariants that keep a system honest once real data flows through it.
Assistants and agents that work on your own material: reading documents, pulling out the fields that matter, and answering questions with citations. Built for messy real files, not a clean demo.
The layer underneath: ingestion, modeling, and the quality checks that make the numbers trustworthy, plus the API and webhook wiring that keeps your tools talking to each other.
Full builds, from the first prototype to the version your customers depend on. Backend, interface, and the deployment that keeps it running after handover.
Ghita Mezzour
Founder & CEO, DecisiveAI
Ex-Minister of Digital Transition, Morocco
“Mouhcine delivers. He consistently produces high-quality work on time, no matter the challenge. Independent and self-driven, he takes full ownership of projects and quickly adapts to new technologies and environments.”
Sofia Zouine
Data Manager
Leyton
“I had the pleasure of working with Mouhcine, and I can confidently say he is one of the most organized and reliable professionals I've collaborated with. As a Data Engineer and RPA Engineer, he consistently demonstrates strong technical expertise while balancing it with a structured and methodical approach to work.”
A mix of automations, data platforms, and production software. Here's what each one does, and how it holds up once real data flows through it.
Product · Study app on Google Cloud
I built SkillDags, a study app for people preparing for cloud exams. A subject is not a list of lessons here. It is a map. Every module says which modules you have to finish first, so a module only opens when you are ready for it. Four subjects and 216 modules are live on Google Cloud right now, and nobody typed any of them into a form. An AI agent wrote them. It connects to the app through a set of 24 tools, and it has to follow the same rules a human editor does.
Learn more→
Project · Content automation
I built a pipeline that turns a single draft into a full publication — a finished article, pull-quotes, quote cards, a cover, an infographic, and a social carousel — then ships it to a blog via a CI/CD pipeline and schedules it across Instagram, LinkedIn, and Pinterest. One approval up front; the rest runs on its own.
Learn more→
Project · Agent tooling
Claude Code can read a screenshot and even rough out an image in code — but both are limited, and Gemini's vision and image generation go well beyond them. They just aren't at hand in the terminal, so each project that wants them re-wires the API from scratch. I built one skill that puts Gemini's see and draw behind a single command — a gateway any agent, skill, or workflow can build on, without touching the API again.
Learn more→
Project · AI product photography
Product · Gym management SaaS
Project · RAG backend
R&D · Postgres optimization
Project · Brand systems
Tell me what you're trying to build or fix, and I'll come back with whether I can help and what a first step looks like. No pitch deck, no pressure.
Connect on LinkedInI read every message myself and reply within 2 working days.
