Youssef Elzahar

About Me

Data analyst and python developer at 3Brothers with hands-on experience in Data Sceince and ML engi neering, NLP,LLM fine-tuning,power bi and SQL. Built and deployed end-to-end AI solutions, RAG-based chatbots, and ML APIs. Strong background in data analysis, ERP systems,building Reports by using SSRS ,and backend development using FastAPI, and SQL. Kaggle Notebook Expert with practical industry experience, I am now integrating LLMs with software to build products for 3Brothers and freelancing on Mostaql.

Experience

Data Scientist Freelancer - Mostaql (march 2026- present)

Built Airlines Recommendation System by using ML, Python, Streamlit, got 95 on ROC of Random Forest Algorithm and I got Full Rating

Python Developer and Data Analyst - 3Brothers (November 2025 – Present)

Worked with ERP systems (LN, Odoo) for data extraction, analysis, and reporting.
Designed Power BI dashboards and SQL Server reports for management insights.
Automated KPI reporting pipelines using complex SQL queries.
Collaborated with stakeholders to analyze operational performance metrics.
building apps integrate with llm apis by using fastapi, html, css,js, I built website back fastapi front hrml,js and Integrating with Chatgpt api this website customize for company, I built webiste integrate with whatsapp and chatgpt to summarize visits of sales engineers this visits are sent everday through whatsapp group and The admin going to ui to ask gpt about results of visits everday and whose did visit today

Full Stack Data Scientist Intern — Fixed Solution (July 2025 – Present)

Built AI agents using n8n triggers.
Developed classic ML models and built Docker images.
Deployed APIs using Flask.
Developed RAG chatbot that supports multi domains.
Worked on Advanced Statistics Analysis on Robot Faults dataset and Chocolate dataset.

ML for Data Analysis — NTI Digital Egypt Youth (Dec 2024 – Jan 2025)

Worked on ML models and preprocessing.
Competed in Kaggle competition “Home Credit Risk”.

Freelance Data Analyst — FreelanceYard (Apr 2025 – Present)

Built interactive sales dashboards in Power BI.
Crafted KPIs using DAX (avg. unit price, quantity sold, discount distribution).

Projects

RoboFaults Worked on a robot fault detection dataset to classify normal vs. fault states across six joints. Identified temperature and current as the most critical factors influencing robot faults [GitHub]
chocolate_factory_analysis This project utilizes two primary datasets Demographic Data (customer attributes) and Purchasing Data (incidence, prices, promotions, and brand choice) to provide a comprehensive market analysis. [GitHub]
Nurse Robot (Graduation Project) Robot to assist doctors during COVID using QR and face recognition. Website + mobile app for control. Grade: Excellent.
Medical Chatbot (LLM Fine-Tuning) Fine-tuned TinyLlama-1.1B on HealthCareMagic-100k using LoRA. Deployed with Gradio for real-time Q&A. [GitHub]
Autocomplete System LSTM-based autocomplete with NLP preprocessing. Also tried GloVe embeddings. Frontend in Streamlit. [GitHub]
Chatbot RAG System Multi-domain chatbot with vector DB embeddings, chat history, Flutter frontend + FastAPI backend. [GitHub]
Real Estate of Egypt BPerformed EDA and KMeans clustering to segment property types; built Power BI dashboards and Streamlit app. [GitHub]
Tech Job Salary Analysis: Prediction & Clustering Built an end-to-end ML pipeline to predict tech salaries and segment professionals using clustering tech niques. Performed EDA, feature engineering, and trained models (Linear Regression, Random Forest, XG Boost) achieving strong predictive performance. Applied K-Means clustering to identify career archetypes (early, mid, senior, business roles). Developed an interactive Power BI dashboard with DAX measures to visualize salary trends across roles, industries, and locations. [GitHub]

About Me

Education

Bachelor of Computer Science (2018 – 2022)