AI/ML Engineer · LLM Agents · End-to-End Solutions
I build AI systems from the business need to production.
Final-year engineering student at ESITH Casablanca. I design and ship LLM agents, machine-learning pipelines and the full-stack systems that run them — currently building an end-to-end AI agent platform at EHC Groupe. Business & Data Management background, so I start from the problem rather than the model.
Available for an end-of-studies internship (PFE) from February 2027
The best vibe coder isn't the one who writes every line manually, but the one who knows the right questions to ask—and the right answers to verify.
01 — About
From the real need to the running system
I build AI/ML solutions end to end — starting from the actual business need and finishing in production. In practice that means LLM agents driven by careful prompting, machine-learning pipelines that monitor and retrain themselves, and the APIs, data models and interfaces that let real users act on the output.
My degree sits between industrial engineering and data: operations and supply chain on one side, machine learning, information systems and generative AI on the other, after two years of CPGE MPSI/MP. That mix is why I care as much about reliability, explainability and testing as about model accuracy — a prediction nobody trusts is not a delivered system.
Civilization advances by extending the number of important operations which we can perform without thinking about them.
02 — Focus
What I work on
LLM Agents & GenAI
Agents built on the Claude API, prompt engineering, RAG, and orchestration with n8n.
Machine Learning
Forecasting, classification and anomaly detection with scikit-learn, XGBoost and PyTorch.
MLOps
MLflow tracking and registries, Airflow scheduling, drift detection, model monitoring, Docker.
Full-Stack Delivery
FastAPI and Express services, Next.js/React interfaces, MongoDB and PostgreSQL, REST APIs.
Knowing is not enough; we must apply. Willing is not enough; we must do.
03 — Selected work
Projects
AI Automation Platform
EHC Groupe · privateInternal platform work combining LLM agent integration with full-stack delivery — from the data model and services through to the interface. Built with the Claude API and orchestrated with n8n.
Further detail is covered by confidentiality.
Smart Supply Forecaster
MLOpsClosed-loop continuous-training pipeline for demand forecasting, built so that model updates are automatic but never blind.
- FastAPI serving with prediction logging.
- Airflow monitors a 14-day rolling RMSE and triggers retraining.
- Chronological validation with no data leakage.
- Promotion audit — a new model ships only if it beats the one in production.
Athena Core
Decision supportDecision-support platform that turns raw business data into strategy, layer by layer.
- Revenue forecasting with Random Forest — R² ≈ 0.92 on the test set.
- Anomaly detection with Isolation Forest.
- Recommendation engine, RFM segmentation and KPI tracking.
- Auto-generated strategic frameworks.
AI Compass
Full-stackDeployed web app comparing large language models on published benchmarks and pricing — MMLU, HumanEval and GSM8K alongside production cost analysis.
- Relational data model with a REST API supporting filtering and sorting.
- Responsive React dashboard.
04 — Experience
Where I have worked
Software Development Intern
Internal AI automation platform
- Building an internal platform that integrates an LLM agent, using prompt design and the Claude API, orchestrated with n8n.
- Responsible for delivery across the stack — data model, backend services, real-time updates and front-end interface — including reliability and testing.
- Stack: Next.js/React, Node.js/Express, Python/FastAPI, MongoDB, Docker.
AI Automation Consultant
- Mapped an end-to-end business process and authored the functional specification that became the blueprint for the platform.
Assistant Process Engineer Intern — Data Science
- Collected and preprocessed industrial data.
- Developed and evaluated machine-learning classification models to support process optimization.
05 — Résumé
Education & skills
Education
-
State Engineering Degree — Industrial Engineering
Business & Data Management major · ESITH, Casablanca
Machine learning, deep learning, data science & AI, generative AI, Big Data, Python (3 levels), databases, cloud computing & cybersecurity, architecture & ERP.
- CPGE — Preparatory Classes for Engineering Schools Mathematics & Physics (MPSI / MP) · Centre Acharif Al Idrissi, Taza
Skills & tools
AI & machine learning
Python, scikit-learn, XGBoost, PyTorch, model evaluation & monitoring, drift detection, MLflow, Airflow, LLM agents, generative AI (Claude API), prompt engineering, RAG
Development
FastAPI, Node.js/Express, Socket.IO, Next.js/React, TypeScript, REST APIs, MongoDB, PostgreSQL, SQL, Docker, Git
Data
Data collection, preprocessing & analysis, feature engineering, pandas, NumPy, Power BI, dashboards & KPIs
Languages
Arabic (native) · French (fluent) · English (fluent) · Spanish (basic)
06 — Credentials
Certifications & leadership
-
Workflow Automation with n8nDataCamp
-
McKinsey Forward ProgramMcKinsey & Company · 2025
-
Project InitiationGoogle
-
UX DesignGoogle
-
Lead, Prospecting Unit — ESSOR Junior EntrepriseJun 2025 – Present
Perfection is attained not when there is nothing more to add, but when there is nothing more to take away.
07 — Contact
Get in touch
Looking for an end-of-studies internship (PFE) from February 2027 — flexible duration, open to opportunities in Morocco and internationally.