Data Scientist & AI Specialist

Hello, I'm Nada

I transform complex data into strategic insights using Generative AI and LLMs. Specialized in audit automation and financial analytics.

LinkedIn Profile
Nada Hammami

About Me

I'm a motivated and dedicated Data Scientist with solid expertise in data science applied to audit and finance. My passion lies in leveraging Generative AI and Large Language Models to build innovative solutions.

Freshly graduated with a strong track record of excellence in both academic pursuits and professional internships across leading tech companies.

I excel at transforming complex data into actionable insights that drive strategic decision-making and process optimization.

Location

Paris, France

Education

National Engineering Diploma in Intelligent and Complex Systems

INSAT (2020 - 2025)

Languages

English C2French C1Arabic NativeGerman B1

Professional Experience

Data Scientist

@Supervizor
Aug 2024 - Present

Paris, France

  • Leading data science initiatives and AI-powered solutions for audit automation and compliance
  • Developing advanced machine learning models for financial data analysis and anomaly detection
Machine LearningLLMPythonGoogle Cloud

Data Scientist Intern

@Supervizor
Feb 2025 - Jul 2025

Paris, France

  • Realized two controls: one for obsolete account detection improving accounting data management, and another for identifying concealed fixed assets in expenses, optimizing compliance and asset tracking
  • Deployed LLM server and configured Datadog for monitoring
  • Developed intelligent system powered by LLM: From OCR Analysis to Text-to-SQL Generation for Compliance Test Automation
LLMOCRText-to-SQLDatadogPython

Data Engineering Intern

@Sofiatech
Aug 2024 - Sep 2024

Tunis, Tunisia

  • Designed and implemented clickstream data analysis pipeline on AWS, with data processing and integration via AWS Cloud9
  • Transformed data and used Amazon CloudWatch Logs Insights to extract relevant information
  • Created interactive dashboards and visualizations for actionable user behavior insights
AWSCloud9CloudWatchPythonETL

Data Scientist Intern

@WaterSec (by Istidama)
Jun 2024 - Jul 2024

Tunis, Tunisia

  • Improved predictive model accuracy by 15% and reduced overfitting by 10% for water consumption pattern recognition using advanced preprocessing, cross-validation, and feature selection techniques
  • Increased user satisfaction by 10% and reduced water consumption by 15% by applying data analytics to optimize facility usage in gym showers
MLPreprocessingCross-validationPythonScikit-learn

AI Engineer Intern

@CodeClause
Aug 2023 - Sep 2023

Tunis, Tunisia

  • Developed handwritten digit and gesture recognition projects using OpenCV, MediaPipe, PyAutoGUI, and PyTorch
  • Implemented computer vision and deep learning solutions for real-time gesture recognition
OpenCVMediaPipePyTorchDeep Learning

Data Scientist Intern

@Tunisair
Jul 2023

Tunis, Tunisia

  • Optimized aircraft energy efficiency while reducing costs by 15% using predictive modeling with linear regression and decision trees
  • Improved operational decision-making by 10% using EDA and data visualization with Matplotlib and Seaborn
Predictive ModelingEDAMatplotlibSeaborn

Education

Engineering Cycle in Intelligent and Complex Systems

@INSAT
2022 - 2025

Tunis, Tunisia

Integrated Preparatory Cycle: Mathematics-Physics-Computer Science

@INSAT
2020 - 2022

Tunis, Tunisia

2020

Béja, Tunisia

High honors (17/20)

Projects

Advanced Search & Recommendation Engine for E-commerce

E-commerce system with content-based recommendations using TF-IDF, word embeddings, and deep text search

  • Created a content-based recommendation system using TF-IDF, word embeddings, PyTerrier, Sentence-BERT, and Deep Text Search
  • Generated personalized recommendations by exploiting user behaviors and product attributes, while providing accurate and multilingual search results
TF-IDFEmbeddingsPyTerrierBERTPython

Tennis Analysis System

Video analysis system for measuring player velocity, ball speed, and shot count using computer vision

  • Implemented a video analysis system using Python, YOLOv8, and CNNs to measure player velocity, ball velocity, and shot count
  • Fine-tuned player, ball, and court key point extraction models using PyTorch, Pandas, NumPy, and OpenCV
YOLOv8CNNPyTorchOpenCVComputer Vision

Tweet Sentiment Classification System

Multi-model sentiment classification achieving 83.6% F1 score on tweet data

  • Developed a tweet classification system using various models such as LSTM, BERT, RoBERTa, and ConvBERT
  • Optimized performance with BERTweet to achieve an F1 score of 83.6%
NLPBERTLSTMRoBERTaTransformers

More projects available on my GitHub profile

SKILLS

With a knack for quick learning, I focus on mastering many skills and technologies needed for Data Science and AI.

Python

Python

Advanced

SQL

SQL

Advanced

R

R

Proficient

Java

Java

Proficient

CONTACT ME

Have a question or want to work together? Feel free to reach out!