AI Engineer : You build it, it ships to production
You are the builder. RAG, multi-agent systems, orchestration: you build inside our clients' tools, and what you write ships to production.
What you take on
You wire an agent into a system that was never designed for it
A client CRM or ERP has ten years of history and empty fields where the agent expects clean data. You decide whether to fix it upstream on the client side or work around it in code, and you document why.
You write code that ships to production, not a demo
A RAG pipeline, a multi-agent system or orchestration, on the Claude, Claude Code, LangChain, Python, TypeScript stack already in place at AI Makers. What leaves your machine runs in front of real users the following week.
You write evaluations before adding a feature
An LLM system never returns the same answer twice: without a test set and regression thresholds, nobody knows whether a change improved or broke the system. You are the one who puts those guardrails in place.
You watch the bill before it blows up
A system that works in a demo and costs 40% more than planned in production does not work. Cost control, model choice, caching: that is your problem, not the client's.
You are on call when it breaks
A production system goes down on a Tuesday evening: you are the one who gets called, and you are the one who fixes it.
You train while building
Part of your time goes to explaining to the client's non-technical teams what the system does and how to adjust it.
Weeks 1-2, you take over a system already in production and evolve it. From month 2, you carry your own workstreams end to end, under the technical direction of Walid Boulanouar, our CTO. In month 3, you deliver at a client directly.
Ce qu'on cherche
- Languages
- Solid Python and TypeScript, comfortable SQL
- Data
- Able to read an existing schema and turn it into a usable RAG corpus
- Architecture
- REST APIs, CRM/ERP integrations, cloud (AWS, GCP or Azure)
- Applied AI
- RAG, multi-agent, orchestration, evaluation, cost control, Claude/Claude Code/LangChain/n8n stack
- Field work
- Has already put a system into production and maintained it, writes clearly
- Background
- 2 to 8 years in software development, part of it on AI systems
Ce qu'on ne cherche pas
Someone who wants to do research. Here we deliver at clients, every week.
Where it differs from neighbouring roles
| Role | What they do | Who owns the outcome |
|---|---|---|
| ML Research Engineer | Trains models, long-cycle R&D | The lab owns the outcome |
| Generalist backend developer | Classic business services, deterministic logic | The product team owns the outcome |
| Pre-sales AI consultant | Recommends, documents, leaves | The client owns the outcome alone once they leave |
| AI Engineer, AI Makers | Builds inside the client's tools, on-call included | You own the outcome until it runs |
Remote, in a time zone compatible with Paris and Casablanca (Europe, North Africa or nearby), with no requirement to live in either city, and occasional time in our offices.
Is this role for you?
No cover letter. The form takes five minutes and starts with showing what you have built.
Apply for AI Engineer