About Essohanam Josue
Ingénieur Data & IA en dernière année à l'ENSA Al Hoceima,
je transforme des données brutes en décisions business claires
et actionnables.
Ce que je livre concrètement :
→ Modèles prédictifs avec recommandations métier (churn,
scoring financier, prévision de demande)
→ Pipelines analytiques end-to-end prêts pour la production
→ Dashboards exécutifs orientés décideurs (Power BI, Plotly)
→ Systèmes RAG et agents IA sur vos documents métier
Résultats récents :
- Analyse de churn pour un fournisseur d'énergie (framework
BCG) — XGBoost Précision = 0.96, recommandation de rétention
ciblée livrée
- Scoring de santé financière pour 9 600 PME en Afrique
Australe — F1 = 0.82, API de scoring déployée
- Prédiction de la demande internet nationale au Togo —
ROC-AUC = 0.86, guidant les investissements infrastructure
2 stages en IA appliquée (ISICOD · 3D Smart Factory) et une
exposition directe aux frameworks consulting BCG.
Je pense comme un consultant et je construis comme un
ingénieur. Je ne m'arrête pas au modèle — je livre la
recommandation.
Disponible immédiatement · Casablanca
French
Native or bilingual
English
Conversational
Remote only
Primarily works remotely
Experience
- GeteamData & Ai EngineerTECHMarch 2026 - Today (3 months)Casablanca, MoroccoDesigning an end-to-end AI system to automatically identify jobs of equal value within companies, ensuring compliance with the EU Pay Transparency Directive 2023/970.The system processes raw HR data through a 6-layer pipeline .
- Fine-tuned CamemBERT for gender bias detection with contrastive learning, built semantic ESCO occupation mapping via FAISS + sentence-transformers + LLM reranking;
- Developed a knowledge graph in Neo4j linking 300+ French collective agreements (CCN) to the EU ESCO taxonomy;
- Designed a Constitutional 3-agent LLM scoring system evaluating 16 sub-criteria with full audit trail;
- Built interpretable clustering with SHAP explainability, producing court-defensible equal-value groups.
- ISICODData & Ai EngineerTECHJune 2025 - September 2025 (3 months)Rabat, Morocco
- Co-built an end-to-end biometric AI platform for fingerprint classification (12 classes) and ridge counting — processing 15,848 images from 5 heterogeneous sources. Produced the largest labeled dataset of its scope in the literature.
- Delivered a production REST API (FastAPI + React) with real-time inference, model performance dashboard, and admin monitoring directly operational for the product team.
- Sole implementer of a Human-in-the-Loop retraining system: user corrections trigger incremental PyTorch retraining, enabling continuous model improvement without static dataset dependency.
- 3D Smart FactoryMachine Learning EngineerTECHJune 2024 - August 2024 (2 months)Mohammédia, Morocco
- Designed custom Kolmogorov–Arnold Network (KAN) layers integrated into a 3D U-Net for volumetric medical image segmentation achieving +3% accuracy gain over the standard 3D U-Net baseline.
- Optimized architecture through systematic tuning of depth, kernel size, and activation strategies using PyTorch on MRI datasets.
Recommendations
Be the first to recommend Essohanam Josue
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- IngénieurEcole Nationale des Sciences Appliquées de Hoceima2026Data engineering
- CPI , Sciences de l'ingénieurÉcole Nationale des Sciences Appliquées de Fès2023CPI , Sciences de l'ingénieur
Certifications
- BCG - Data Science Job SimulationForage2026