Amine El Faiz, Developer in Rabat, Rabat-Sale-Zemmour-Zaer, Morocco
Amine is currently unavailable

Amine El Faiz

Data Science Developer

Rabat, Rabat-Sale-Zemmour-Zaer, Morocco

Toptal member since March 7, 2022

Bio

Amine is a highly skilled data scientist and AI engineer with over seven years of experience driving impactful business outcomes through AI and machine learning (ML). With a proven track record across different sectors, he has developed and deployed predictive algorithms and AI systems that enhance customer engagement, optimize processes, and increase revenue. Amine is proficient at learning new skills and loves to work on challenging projects that create value.

Portfolio

Ship Angel, Inc.
Python, SQL, Pandas, Microsoft Excel, Amazon Web Services (AWS), AI Agents...
Legacy South Builders, LLC
Machine Learning, Data Science, Retrieval-augmented Generation (RAG)...
Artefact
Dialogflow, Vertex AI, Google Cloud Platform (GCP), LangChain, Chatbots...

Experience

  • Data Science - 6 years
  • Python - 6 years
  • Machine Learning - 6 years
  • Pandas - 5 years
  • Statistics - 5 years
  • Deep Learning - 4 years
  • Natural Language Processing (NLP) - 3 years
  • AI Agents - 2 years

Preferred Environment

Jupyter Notebook, Python, Machine Learning, Data Science, Amazon Web Services (AWS), Google Cloud Platform (GCP), Git, Artificial Intelligence (AI)

The most amazing...

...experience I've had was working with international consultants and customer service teams to increase the bank's revenue generated from personal loans by 27%.

Work Experience

Data Scientist/Engineer

2024 - PRESENT
Ship Angel, Inc.
  • Built AI agents and RAG systems that are connected to multiple sources of information related to shipping and can answer clients' questions from those sources.
  • Automated the process of extraction rates, surcharges, and all information related to shipping costs from different file types (Excel, PDFs, etc.) using AI, AWS, Python, and OCR. The pipeline reduces manual processing by 50%.
  • Developed MLOps pipelines using Sagemaker to automate the training and inference of ML and NLP models.
Technologies: Python, SQL, Pandas, Microsoft Excel, Amazon Web Services (AWS), AI Agents, Retrieval-augmented Generation (RAG), Artificial Intelligence (AI)

Data Science & AI Expert | Real Estate Solution

2025 - 2025
Legacy South Builders, LLC
  • Developed an automated workflow that replaced the manager’s manual task of monitoring construction schedules by detecting discrepancies and gathering delay reasons through AI, resulting in faster reviews and more reliable project tracking.
  • Deployed an AI-driven voice agent that contacted workers directly, collected explanations for schedule delays, and fed their responses back into project schedules for real-time updates.
  • Delivered and deployed the first version of the solution in under five weeks, providing immediate business value by reducing delays, improving data accuracy, and streamlining communication between managers and field teams.
Technologies: Machine Learning, Data Science, Retrieval-augmented Generation (RAG), Vector Databases, Artificial Intelligence (AI), Python, Data Lakes, ETL, n8n, Retell AI, Google Cloud Platform (GCP), Smartsheet

AI Engineer

2023 - 2024
Artefact
  • Developed a generative AI agent on Google Cloud Platform (GCP), utilizing retrieval-augmented generation (RAG) to answer customer inquiries from the client website as well as from additional resources like documents and FAQs.
  • Built a large language model (LLM) evaluation pipeline for hyperparameter tuning and a monitoring system to ensure continuous, accurate performance of the chatbot responses.
  • Achieved a 30% reduction in customer handovers to human assistance, enhancing customer satisfaction and operational efficiency.
Technologies: Dialogflow, Vertex AI, Google Cloud Platform (GCP), LangChain, Chatbots, Large Language Models (LLMs), Artificial Intelligence (AI), Retrieval-augmented Generation (RAG), Google Cloud Functions, Vector Databases, Training, Training Workshops, AI Agents

Senior Data Scientist

2022 - 2023
Forbes LLC - Main
  • Developed ML algorithms that accurately forecast the churn behavior of Forbes subscribers, leveraging a comprehensive dataset encompassing subscription details and browsing activity, such as session timestamps, time on page, and additional metrics.
  • Designed machine learning models that predict the anticipated advertising revenue generated from publishing a specific article on the Forbes platform.
  • Performed in-depth analysis to identify the most important features influencing model predictions, leading to better insights and driving business growth.
Technologies: Python, SQL, Machine Learning, Natural Language Toolkit (NLTK), TensorFlow, Keras, Gensim, Google Cloud Platform (GCP), Shapely, Pandas, Jupyter Notebook, Generative Pre-trained Transformers (GPT), Natural Language Processing (NLP), Artificial Intelligence (AI), Text Classification, Classification, Data Analytics, Data Visualization, Data Analysis, Predictive Modeling, Vertex, Google Cloud, Google Cloud Machine Learning, Jupyter, Named-entity Recognition (NER), Product Management, Data Manipulation, Analytics, Leadership

Data Scientist

2021 - 2022
INWI
  • Developed machine learning models to predict the churn of prepaid clients trained on historical data such as recharges, internet consumption, and incoming and outgoing calls.
  • Clustered prepaid clients into business segments for model evaluation and better customer retention campaigns.
  • Estimated the potential return of investment from predictive models.
Technologies: Jupyter Notebook, Python, Scikit-learn, Pandas, Computer Science, Machine Learning, Data Science, SQL, Critical Thinking, Statistics, NumPy, Artificial Intelligence (AI), Classification, Arabic, Predictive Modeling, FastAPI, Jupyter, Analytics

Data Scientist

2018 - 2021
Banque Centrale Populaire
  • Developed machine learning models that predict potential clients needing mortgage and personal loans in the future.
  • Tested ML algorithms in the market and estimated the potential return of investment from the results.
  • Deployed solutions into production and automated the process of training models and scoring clients.
  • Developed a chatbot that can answer all customers' questions related to mortgages and help them calculate their monthly payments.
  • Programmed an optical character recognition algorithm that extracts text from ID card images (name, date of birth, address, etc).
  • Collaborated with an international consulting firm to create a data-driven culture through presentations and peer-to-peer learning.
  • Led the interview process of recruiting data scientists and managed the data science team.
Technologies: Scikit-learn, Pandas, Git, SQL, Python, Deep Learning, Jupyter Notebook, Computer Science, Machine Learning, Agile, Data Science, Data Visualization, Tesseract, Keras, Docker, Random Forests, A/B Testing, Convolutional Neural Networks (CNNs), Generative Pre-trained Transformers (GPT), Natural Language Processing (NLP), XGBoost, Logistic Regression, Critical Thinking, Statistics, OpenCV, Word2Vec, Optical Character Recognition (OCR), NumPy, Artificial Intelligence (AI), Text Classification, Classification, Data Analytics, Data Analysis, Arabic, Predictive Modeling, Jupyter, Finance, Trend Analysis, Computer Vision, Team Leadership, Product Management

Experience

Prepaid Churn Prediction

A machine learning model that predicts the prepaid clients that will leave a telecommunication company in the future.

• Data analysis (recharges, internet consumption, calls, etc.) to characterize churners versus active clients.
• Formulation of a machine learning problem that solves the underlying business issue.
• Data preparation and feature engineering.
• Model building and evaluation with clear business metrics.
• Error analysis to enhance model performance.
• Model interpretation to understand client churn reasons.

The model was ten times better at detecting churners than random selection. To retain clients, targeted offers were sent based on purchase history and were able to reduce the number of churners by 20%.

Client Loan Prediction

A machine learning model that predicts clients' intentions to acquire personal and mortgage loans in the future.

• Data definition, extraction, and storage in PostgreSQL tables (static data, transactions, products, loans, etc).
• Model building and market evaluation using A/B testing.
• Model deployment and automation of training and scoring pipelines.
• Estimation of expected annual revenue from the predictive model.

Chatbot

A deep learning algorithm that answers clients' questions related to loans (interest rate, loan eligibility, loan documents) and simulates the amount that will be paid monthly by clients wanting to have a mortgage.

Education

2015 - 2018

Engineer's Degree in Industrial Engineering

Mohammadia School of Engineering - Rabat, Morocco

2013 - 2015

Higher School Preparatory Classes in Mathematics and Physics

CPGE Mohammed V - Casablanca, Morocco

Certifications

APRIL 2023 - PRESENT

AWS Cloud Practitioner Essentials

Coursera

MAY 2020 - PRESENT

Deep Learning Specialization

DeepLearning.AI

Skills

Libraries/APIs

Scikit-learn, Pandas, XGBoost, Natural Language Toolkit (NLTK), NumPy, Keras, OpenCV, TensorFlow, Shapely, PaddleOCR

Tools

Jupyter, Git, Claude, Plotly, Seaborn, Gensim, LayoutLM, Named-entity Recognition (NER), Tableau, Dialogflow, n8n, Retell AI, Smartsheet, Microsoft Excel

Languages

SQL, Python, Arabic

Platforms

Jupyter Notebook, Google Cloud Platform (GCP), Docker, Amazon EC2, AWS IoT, Amazon Web Services (AWS), Vertex AI, AWS Lambda

Storage

Google Cloud, PostgreSQL, Data Lakes, Redshift

Frameworks

LangGraph

Paradigms

Agile, ETL

Other

Machine Learning, Data Science, Programming, Critical Thinking, Communication, Deep Learning, Natural Language Processing (NLP), Supervised Machine Learning, Artificial Intelligence (AI), Code Review, Task Analysis, Technical Hiring, Classification, Data Analysis, Predictive Modeling, Large Language Models (LLMs), Product Management, Retrieval-augmented Generation (RAG), Mathematics, Computer Science, Statistics, Optimization, Tesseract, Word2Vec, fastText, Genetic Algorithms, Optical Character Recognition (OCR), Interviewing, Generative Pre-trained Transformers (GPT), Text Classification, PDF Scraping, Language Models, LangChain, Hugging Face, Data Analytics, Data Visualization, Vertex, Google Cloud Machine Learning, Real Estate, Computer Vision, Team Leadership, Google BigQuery, Data Manipulation, Analytics, Time Series Analysis, Leadership, OpenAI GPT-4 API, Chatbots, Google Cloud Functions, Vector Databases, Training, Training Workshops, AI Agents, Software Architecture, A/B Testing, Presentations, Scraping, Text Categorization, Random Forests, Convolutional Neural Networks (CNNs), Logistic Regression, FastAPI, Linear Regression, Fine-tuning, Prompt Engineering, Label Studio, Finance, Trend Analysis, Layout Parser, Detectron2, APIs

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring