[nom]
[téléphone] | [e-mail] |[lien] | [lien]
EDUCATION
ENSIAS, National School of Computer Science and Systems Analysis · Sep 2023 - May 2026
Engineering Student in Business Intelligence & Analytics
Relevant Coursework: Machine Learning, Statistical Modeling, Data Mining, Database Systems, Big Data Analytics, Decision Support
Systems, SQL, PL-SQL
Semlalia University · Sep 2020 - May 2023
Bachelor's Degree in Mathematical Sciences and Computer Science
Relevant Coursework: Probability, Statistics, Linear Algebra, Calculus, Algorithms, Programming, SQL, PL-SQL
TECHNICAL SKILLS
Programming & Data: Python, SQL, Java, pandas, NumPy, scikit-learn, statsmodels | ML & Statistics: classification, regression, anomaly detection, hypothesis testing, confidence intervals, precision/recall, F1-score, MAE/RMSE | NLP/LLMs: TF-IDF, transformer embeddings | Tools: Selenium, Git, Matplotlib, Seaborn, Plotly, SQL Server
WORK EXPERIENCES
KShuttle Maroc · Casablanca
Data Sustainability & Analytics Intern | Python · pandas · Data Modeling · ESG Analytics · Feb 2026 - Present
- Designed ETL pipelines and analytical data models for sustainability/SSE reporting, translating regulatory requirements into measurable indicators, data fields, and validation rules.
- Cleaned and structured ESG-related datasets, checking missing values, inconsistencies, and reporting logic across source systems to support compliance analysis.
- Used pandas, NumPy, scikit-learn, and visualization libraries to explore patterns and identify anomalies in structured sustainability data.
AIBOOST for Industries · Casablanca
Data Scientist | Python · NLP/LLM · Data Pipelines · Insight Extraction · Sep 2024 – Sep 2025
- Developed Python data pipelines and Selenium web scraping workflows to collect, clean, and structure large-scale online and internal datasets
- Processed call transcripts with NLP/LLM workflows to extract recurring themes, customer pain points, and business insights into structured JSON/PDF reports.
- Validated generated insights against source transcripts and structured outputs to improve reliability.
PROJECTS
AI CV Screener | RAG-Powered Recruitment Assistant | Python · Streamlit · LangChain · ChromaDB · Groq API ….
- Built an end-to-end RAG pipeline to automatically screen and rank candidates by ingesting a job description and multiple CVs (PDF), indexing them into a ChromaDB vector store using HuggingFace sentence embeddings, and retrieving the most relevant chunks for context-aware evaluation.
- Prompted Llama 3.3 70B via Groq API to score each candidate (0–100), extract structured information (skills, experience, education, languages), and generate a written justification explaining the fit or gaps relative to the job requirements.
- Delivered results through an interactive Streamlit dashboard with a comparative ranking table and exportable PDF/Excel report, reducing manual CV review from hours to seconds.
ESG Anomaly Detection | Python · pandas · scikit-learn · Unsupervised Learning
- Developed an unsupervised anomaly detection workflow to flag outliers, inconsistent values, and potential data-quality issues in ESG datasets without labeled examples
- Prepared features through cleaning, normalization, and exploratory analysis, then interpreted anomalies against distributions and business rules for explainable review.
Conversion & Revenue Drivers Analysis | Python · statsmodels · scikit-learn · Linear Regression
- Analyzed customer/transaction data to identify factors associated with conversion, purchase value, or revenue using EDA, correlation checks, feature preparation, and linear regression baselines.
- Evaluated models with train/test split, MAE, RMSE, and residual analysis; used confidence intervals, p-values, and hypothesis testing to assess meaningful group differences and separate correlation from causation.
LANGUAGES
English: Fluent | French: Fluent | Arabic: Native
