M
mahmoudtrigui's photo
From Tunisia 07:44 PM (GMT+01:00)
$100/hr or $120,000/yr

Active 1 hour ago


Share this profile:

Data Scientist

Data Scientist
Available for hire
Years of experience
10+ years
Experience level
Senior
Available for
Part-time, Contract, Freelance

Data Scientist with 10+ years of experience specializing in structured data analytics/machine learning, time series forecasting, and NLP solutions. With expertise in both R and Python ecosystems.

I am particularly passionate about feature engineering and its crucial role in machine learning effectiveness.

Core competencies include:

▪ Structured tabular data analysis & ML solutions

▪ Time series forecasting

▪ Clustering algorithms

▪ NLP & LLM implementations

▪ Advanced feature engineering

Consistently expanding expertise through competitive programming and continuous learning from

leading educational platforms. Combines technical proficiency with strong analytical thinking to deliver

impactful data-driven solutions.

For more, i invite you to visit my website : https://www.mahmoudtrigui.com/

Employment History

Data Scientist at Sofrecom Tunisia Current 2024 - Now
❖ Time Series Forecasting Framework for Resource Planning ▪ Project Scope ✓ Developed 12-month and 52-week forecasting system for multiple independent time series ▪ Technical Implementation ✓ Built a semi-automated forecasting framework in Python using MLforecast library (lag/features)/model ✓ Implemented comprehensive feature engineering including: (date, Seasonality/ Trend/ Periodicity) ▪ Techniques ✓ Applied target transformation functions for improved accuracy ▪ Client Engagement & Visualization ✓ Facilitated data-driven decision making through PowerBI and LookerStudio ❖ PDF Audit Information Extraction System with LLM ▪ Intelligent Document Processing ✓ Developed automated extraction system for audit reports using GPT-4 ✓ Implemented multi-page PDF processing with image conversion ✓ Built robust validation system for extracted data ▪ Technical Implementation ✓ Designed modular Python architecture for scalable processing ✓ Created custom data validation using Pydantic models ✓ Built automated JSON output generation ❖ FTTH Access Orders Forecast ▪ Project Scope ✓ 12-month forecast horizon (weekly/monthly granularity) ✓ Multi-dimensional predictions: global, departmental, and offer-specific ✓ Limited historical data and feature availability ▪ Technical Implementation ✓ Developed comprehensive feature engineering pipeline ✓ Leveraged advanced time series decomposition techniques ✓ Implemented ensemble modeling approach using R's feasts and fable ✓ Achieved ~5% RMSE through optimized model combination ❖ Co-Financing Prediction Model (Data Pipeline Migration & Production Implementation) ▪ Architecture Optimization ✓ Converted Python scripts to production-grade Dataiku workflows ✓ Prioritized SQL and native Dataiku components for performance ✓ Designed efficient Analytics Base Table (ABT) creation process
Data Scientist at Kiota Intelligence 2021 - 2023
Building Data Science solutions for the VC industry. All projects and productions below were built with R Language ▪ Predictive Models ✓ Startup survival prediction ✓ Series A funding probability estimation ✓ Pre-valuation modeling ✓ Investor-startup matching algorithm ▪ Interactive Analytics ✓ Built Shiny dashboard for funding visualization ✓ Implemented dynamic filtering tools ✓ Developed multidimensional outlier detection system ✓ Created temporal clustering analysis ▪ Automated Reporting ✓ Designed CSS-styled RMarkdown reports ✓ Implemented automated email distribution system ✓ Developed statistical testing framework for funding rounds ✓ Created interactive filtering system for investor matching
Data Analyst/scientist at Tunisia Telecom 2016 - 2020
❖ Customer Segmentation ▪ Data Engineering ✓ Constructed Analytics Base Table (ABT) from multiple data sources ✓ Implemented robust data quality checks and cleaning protocols ✓ Executed advanced feature engineering and transformations ▪ Advanced Analytics ✓ Performed statistical testing and correlation analysis ✓ Developed K-means clustering model for behavioral segmentation ✓ Created detailed segment profiles for business validation ▪ Production Implementation ✓ Designed automated classification system for new customer assignment ✓ Delivered actionable customer insights for business strategy ❖ Multi-SIM User Detection ✓ Developed and validated hypotheses for dual-SIM usage patterns ✓ Engineered comprehensive customer scoring algorithm ✓ Created detailed user profiles for target segment ✓ Implemented scoring model using SAS Guide and Miner ❖ Family Communities ✓ Distinguished household-level connections within larger communities from a predefined communities result ✓ Identified extended family networks through behavioral patterns ❖ Collaborated with SAS, KPMG, and Business&Decision expertsin workshop series for different models including : ✓ Churn, Cross-sell and Community Link Analysis (SAS) ❖ SQL Ad-hoc Request & Dashboard Development ✓ Customer Value Management (CVM) performance ✓ Network quality KPIs & Data service penetration ✓ Customer segmentation profiles ✓ Sales analytics and market growth ❖ Process Automation ✓ Implemented VBA automation for PowerPoint reportings ✓ Established data quality verification protocols ❖ Business Recommendations ✓ Designed targeted marketing campaigns ✓ Created Try & Buy offer frameworks ✓ Generated data-driven product recommendations

Education

Engineer of Statistics and data analysis at Higher School of Statistics and Data Analysis 2011 - 2015