Freelancer profile translated to English.

Back to original language

Description

I put generative AI into production — and make it stick.

Senior infrastructure & SRE engineer (7 years in critical environments: Ministry of

Interior, international luxury, retail, VFX post-production), now specializing in

AI agent orchestration and sovereign AI.

What makes me rare: I don't just talk about agents — I operate a fleet of them in production.

A multi-agent system (OpenClaw, Hermes) manages a real Proxmox infrastructure (17

containers, 3 VLANs) via SSH and MCP, 24/7, with governance (permission matrix, human-in-

the-loop) and LLMs hosted 100% locally (MLX, Ollama, LM Studio; Apple Silicon and AMD GPUs).

From server to agent.

The #1 barrier to AI adoption in business is not technology — it's trust:

how to let an agent touch production without risk, and without sending its data to a

third-party cloud? That's exactly my job.

What I can do for you:

• Move your AI POCs into production: governed agents, RAG, on-premise LLM (sovereignty, GDPR, cost control).

• Objectively choose the right model (reproducible benchmark on your task).

• Integrate AI into your operations (AIOps), leveraging 7 years of monitoring and security.

🛠 OpenClaw · Hermes · MCP · MLX/Ollama/LM Studio · Python (FastAPI, RAG) · Proxmox ·

Prometheus/Grafana/Checkmk · Tailscale

🎯 TARGET MISSIONS

→ Move AI POCs into production (governed agents, RAG, on-premise LLM)

→ Deploy sovereign / local-first AI (sovereignty, GDPR, controlled costs)

→ Benchmark & model selection for your use case

→ AIOps: enrich alerts & operations with AI (Prometheus, Grafana, Zabbix)

→ "Agents-Ready" audit + roadmap

📍 Paris / Île-de-France · hybrid or full remote work · immediate start.

Languages

French
Native or bilingual
Arabic
Native or bilingual
English
Fluent

Workplace preferences

Can work on-site

Paris (up to 50km)

BY RAED
AI Consultant & Agent Orchestration
DIGITAL AND IT
September 2025 - Today (11 months)
Paris, France
Applied R&D and consulting in AI agent orchestration. Design and operation of a multi-agent system in production (OpenClaw, Hermes) managing a Proxmox infrastructure via MCP, with governance (N1-N4 permissions, human-in-the-loop) and 100% local LLMs (MLX, Ollama, LM Studio). Reproducible benchmark harness (6 models). Offering: moving AI POCs to production, securely.
LLMOps MCP Artificial Intelligence LLM Ollama
Conforama
Systems & Automation Consultant
E-COMMERCE
September 2024 - September 2025 (1 year)
Lognes, France
Mission on national retail infrastructure (230 Windows servers, 500+ VMs).
Led the migration of 230 Windows Servers (2012 R2 → 2019/2022) in 6 waves: 180 servers migrated, 98% success rate, average downtime 2h15.
Administration of Centreon on 500+ servers (Windows/Linux/DB/network probes).
Developed the use of the ELK stack for centralized log analysis.
Deployment and administration of the Bastion solution for 500+ VMs. N3 support
ServiceNow. PowerShell automation.
Windows Server VMware vSphere Powershell Elastic Stack (ELK) Bastion Wallix
AMI PARIS
IT Systems & Operations Manager
LUXURY GOODS
June 2023 - April 2024 (10 months)
Paris, France
Scope: 90 global stores (Europe, Asia, USA), 330 users, 50+ virtual servers.
Full build to run of Zabbix 6.4 solution from scratch (templates, Veeam API, Teams webhook) — result: -90% incident detection time. Deployment of Addigy MDM on 150+ iPads across 90 stores with region-specific strategies. Administration of hybrid AD / Azure AD for 330 users. Migration from SonicWall firewall → Sophos.
Build to Run GestSup (ticketing). IT support team leadership.
Zabbix NinjaOne VMware vSphere Microsoft Entra ID Monitoring