Skip to content
View fbenkhelifa's full-sized avatar

Highlights

  • Pro

Block or report fbenkhelifa

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
fbenkhelifa/README.md

Farouk Benkhelifa

AI/ML Engineer · AI Evaluation & Computer Vision · Data & Telecom Systems Engineer

📍 Oran, Algeria · State Engineer graduate in Telecommunications and Digital Technologies from ENSTTIC Oran

I build applied AI systems across the ML lifecycle: data pipelines, model development, evaluation, and deployment, with a strong focus on multimodal AI evaluation, LLM quality assessment, and telecom-aware data systems.


About me

I am a State Engineer graduate from ENSTTIC Oran in Telecommunications and Digital Technologies. My work sits at the intersection of AI engineering, computer vision, data science, and telecom systems. My main project focus is HALO / AIGC quality assessment: evaluating AI-generated images through MOS prediction, prompt-image semantic alignment, and reproducible evaluation protocols.

  • Current focus: AI evaluation, AIGC quality assessment, LLM evaluation, and applied ML systems
  • Strengths: end-to-end ML delivery, experimentation discipline, multilingual AI evaluation, and deployment-aware engineering

À propos (FR)

Je suis ingénieur d’État diplômé de l’ENSTTIC Oran en Télécommunications et Technologies Numériques. Je travaille sur des systèmes d’IA appliquée couvrant les données, la modélisation, l’évaluation et le déploiement, avec un intérêt particulier pour l’évaluation multimodale, la vision par ordinateur et les systèmes télécoms/data.


Skills

  • Machine Learning & AI: PyTorch, TensorFlow, scikit-learn, Hugging Face, LLM evaluation
  • Evaluation & Computer Vision: MOS prediction, prompt-image semantic alignment, multimodal evaluation, reproducible evaluation protocols
  • Data & Streaming: Kafka, Spark, Elasticsearch, Kibana, SQL, pandas
  • Telecom & Systems: networking fundamentals, telecom systems engineering, reliability-focused engineering
  • Development: Python, FastAPI, Django, SQLite, Docker

Featured projects

Project What it delivers
HALO AIGC quality assessment toolkit focused on prompt-image semantic alignment, MOS prediction, and reproducible evaluation protocols.
AI School Website Educational platform project that explores AI and modern web experiences.
StreamSight Streaming analytics prototype for near-real-time processing and monitoring workflows.
Store Manager Product-focused desktop application for inventory and store operations.

Experience highlights

  • LILT AI — LLM evaluation and AI training experience focused on model-quality workflows and training support.
  • Telecom internships — Djezzy and Algérie Télécom, with exposure to network systems and service engineering contexts.
  • Data and streaming systems — hands-on work with Kafka, Spark, Elasticsearch, and Kibana for monitoring and analytics workflows.

Open to

  • AI/ML Engineer roles
  • AI Evaluation / Model Quality Engineer roles
  • Data Engineering / Streaming Systems roles
  • MLOps or deployment-oriented ML opportunities

Connect

Popular repositories Loading

  1. store-management-webapp store-management-webapp Public

    Legacy PHP/MySQL store management prototype; active productization continues in store-manager.

    PHP

  2. ml-notebooks-collection ml-notebooks-collection Public

    Collection of exploratory ML notebooks across tabular, NLP, and computer vision tasks.

    Jupyter Notebook

  3. ai-school-website ai-school-website Public

    Canonical AI School Website repository: multi-agent RAG educational platform architecture and roadmap.

    PHP

  4. binary-prediction-with-a-rainfall-dataset binary-prediction-with-a-rainfall-dataset Public

    Notebook-based rainfall binary classification workflow for applied ML experimentation.

    Jupyter Notebook

  5. predicting-road-accident-risk predicting-road-accident-risk Public

    Kaggle tabular regression pipeline for road accident risk prediction with Optuna, LightGBM, and XGBoost.

    Jupyter Notebook

  6. telecom-customer-churn-ml telecom-customer-churn-ml Public

    Telecom customer churn prediction notebook using IBM Telco dataset with recall-focused modeling.

    Jupyter Notebook