Akharaz Jihane

Engineering Student · Statistics & Big Data · INSEA

Engineering student at INSEA, specialising in Statistics and Big Data. Passionate about quantitative analysis and predictive modelling, open to opportunities where technical skills meet strategic consulting and data-driven decision support.

Education

Institut National de Statistique et d'Économie Appliquée (INSEA)
Engineering Degree — Statistics, Demography & Big Data
Classes Préparatoires aux Grandes Écoles (CPGE)
Mathematics, Physics & Engineering Sciences
Baccalauréat — Sciences Mathématiques A
Graduated with Distinction (Très Bien)

Experience

Ministry of Economy and Finance — DEPF
Intern — Macro-Financial Analyst
  • Market Analysis: Studied the impact of Bank Al-Maghrib monetary policies and the Casablanca Stock Exchange performance on the national economy.
  • Modelling & Synthesis: Analysed interconnections between money and capital markets to identify economic growth levers.

Projects

Banking Customer Experience Intelligence Platform – Morocco
  • Built an end-to-end AI pipeline scraping 1,500+ Google Maps reviews (8 banks, 8 cities) via Playwright + Airflow, transformed with DBT, and enriched with multilingual BERT sentiment analysis (88.4% accuracy).
  • Deployed a RAG chatbot (ChromaDB) and autonomous multi-tool agent (Groq LLaMA-3.3-70b) generating consulting reports on Chainlit. Key finding: 78% negative reviews – waiting time & staff as top drivers.
Stress Test Solvency II — MASI/OAT Risk Assessment (Morocco)
  • Modelled market dependence using Gaussian Copulas & EVT-GPD; demonstrated that standard normal distributions underestimate tail risk by 40%.
  • Calculated Solvency Capital Requirement (SCR) under equity (-49%) and rate shocks (+42%), identifying a 54M MAD deficit.
  • Optimized asset allocation using Random Forest + SHAP, recommending a 25% MASI exposure (a 10x more cost-effective solution than capital raising).
Stochastic Mortality Forecasting & Longevity Risk Quantification
  • Benchmarked 5 actuarial models (Lee-Carter, CBD, Kalman) on HMD France data (1950–2024). Validated via rolling backtest (RMSE: 0.188 yr).
  • Quantified longevity risk under Solvency II: recommended a €595K provision (4.1% of capital) for a €14.37M annuity portfolio.
  • Developed a real-time Streamlit monitoring dashboard and a modular Python CLI pipeline for automated reporting.
Geospatial Analysis Dashboard — Power BI & Google Earth Engine
  • Extracted urban and environmental indicators via Google Earth Engine APIs for spatial trend analysis.
  • Designed an interactive Power BI dashboard for cartographic visualization and executive reporting.

Technical Skills

Data Science & AI
Python (Scikit-Learn, TensorFlow) NLP (BERT) Time Series (ARIMA, GARCH) Explainable AI (SHAP)
Actuarial & Risk
Solvency II (SCR, VaR) Extreme Value Theory (EVT) Copulas Lee-Carter/CBD Models Monte Carlo
Data Engineering & BI
SQL (PostgreSQL) Airflow DBT Streamlit Power BI Git

Languages

Arabic — Native French — Professional English — Fluent

Key Strengths

Analytical Thinking Technical Communication Methodological Rigour Operational Adaptability