Freelancer profile translated to English.

Description

Do you want to integrate concrete AI solutions into your products or automate complex business processes?

I help companies design and deploy LLM-based systems (GPT, Llama, Mistral...), particularly for use cases like document analysis, workflow automation, or intelligent assistants.

💡What I can bring you:

Automation of document processing (contracts, invoices, business data)
Implementation ofRAGpipelines for effective use of your internal data
Design of AI agents (**multi-agents**) to orchestrate complex tasks
Model adaptation via fine-tuning (**SFT, RLHF**) for specific use cases
Deployment of scalable and robust cloud solutions (**Azure, Amazon**)

🚀My added value:

Experience in real-world deployment (not just POCs)
Expertise across the entire chain: from data to deployment
Ability to transform a business need into a concrete AI solution
Specialization in complex cases (legal, reasoning, multi-task)

📦Types of projects I undertake:

Document analysis systems (legal, financial...)
Advanced conversational agents (RAG / multi-agents)
Data processing and enrichment pipelines
Integration of AI into existing products (API, backend, SaaS)

🛠️Main Stack:Node.js, Python, LangChain, LangGraph, Azure, vector databases (FAISS, ChromaDB), LLMs (OpenAI, Claude, Mistral…)

💬 Available to discuss your needs and quickly identify high-value use cases.

Industry field of expertise

Languages

French
Native or bilingual
Arabic
Native or bilingual
English
Native or bilingual

Workplace preferences

Can work on-site

Paris (up to 50km)

Contractzlab
AI / ML Engineer
December 2023 - Today (2 years and 6 months)
Paris, France
Design and deployment of amulti-agentlegal AI platform in production, based onLangChainand **LangGraph**, orchestrating specialized agents for contract analysis and regulatory assessment.
Implementation of complete fine-tuning pipelines onDatabricks**, using **SFT(Supervised Fine-Tuning) andRLHFapproaches, applied toLlama**, **Mistral**, **Qwenmodels as well as proprietary models, to adapt general LLMs to complex legal use cases.
Development of hybridRAGpipelines integratingChromaDB**, **FAISS**, and **Azure AI Searchas vector databases, enabling precise search over large legal corpora (contracts, regulations, forms, court decisions).
Construction of a multi-tasklegal benchmarkdataset (several thousand annotated examples), structured by task type (extraction, classification, reasoning...), allowing rigorous evaluation ofLLMmodels.
Production deployment of AI models and agent pipelines on **Azure**, enabling complete automation of complex legal document processing at scale.
LLM AI Agent RAG Python Node.js
SupplyzPro
DevOps / Backend Engineer
October 2023 - December 2023 (2 months)
Tunis, Tunisia
Design of amicroservicesarchitecture deployed withDocker**Swarm**, including load balancing, auto-scaling, and high availability of containerized services.
Implementation of aCI/CDpipeline with **GitLab**, automating build, test, and deployment phases, significantly reducing manual operations.
Deployment of a monitoring stack based onGrafanaandPrometheusto supervise container health, identify bottlenecks, and ensureSLAcompliance.
Docker Kubernetes Gitlab CI/CD Microservices Grafana
Novaspace Munich
Full-Stack Developer
June 2022 - September 2023 (1 year and 3 months)
Germany
Development of an intelligent receipt processing pipeline based onAzure Document Intelligenceand **LLMs (GPT)**, enabling automatic extraction of structured data (supplier, amount, date, category) and expense classification, integrated into an expense management platform.
Design of a financial expense management tool in **React**, **Azure Functions**, and **Express.js**, facilitating data-driven decision-making.
Development of a dataAPIforPower BIvia **Azure Functions and Azure SQL**, enabling data management, transformation, and feeding for analytical dashboards.
React.js LLM Python Node.js Microsoft Azure