Hachem SFAR

Involving in the development of specialized Kafka connectors for SAP technologies, with a focus on coding in Scala and a bit of Java. My work involved developing SMT, providing support to clients who were using our connectors to solve issues they encountered, and working on the Confluent platform to set up connectors. Additionally, contributing to the writing of connector documentation and utilized SonarQube and Jenkins to ensure that the connector was working correctly and free from bugs. Keywords: Scala, Java, Kafka Connect, SAP Technologies, Confluent Platform, SMT, SonarQube, Jenkins

Collaborating with Tomorrow University of Applied Sciences and KfW to analyze a historical dataset and predict the number of new charging stations that would be required for each German state in the coming years. Based on our analysis, we provided recommendations to KfW on which cities and companies would benefit the most from increasing their investment in charging stations. Keywords: Python; Data preprocessing (Pandas, NumPy); Data exploration (Matplotlib, folium); Machine Learning (NeuralProphet, DBScan, HDBScan, Logistic Regression/SVC, KNN); Web Application (Streamlit)

My responsibilities included creating course materials and lessons, teaching groups of students, assessing student work, and collaborating with a diverse and multidisciplinary teaching team. I also coordinated with the ReDI team to improve the courses and the student and teacher experience. The course covered topics such as Python Libraries and Frameworks (NumPy, Pandas, Seaborn), Basic Statistics, Data preprocessing, Analysis, and Visualization using Matplotlib, and Machine Learning.

Leading a project to develop a model that locates the helipads, using machine learning techniques such as classification and transfer learning. I also integrated this model into a web application using Streamlit and used GradCAM for model interpretation. Keywords: Python, Machine Learning, Classification, Transfer Learning, GradCAM, Streamlit, Web Application

Development of a highly accurate machine learning model that predicts the stage of Alzheimer's using MRI images. This model helps doctors diagnose patients more accurately and provide better care. Additionally, Development of a first assessment tool that predicts the dementia risk for a patient based on their medical data, even when MRI images are not available. Keywords: Python, Project Management, Machine Learning, Classification, Streamlit

Contribution in multiple projects across a variety of domains, including data analysis, unit testing, NLP, web development, web scraping, software development, and documentation. Keywords: Python, Tkinter, Pandas, Selenium, SQL, PowerBI, Sphinx, Unit testing, Matplotlib, GIT, Azure storage explorer.

Developing a hate speech detection system using various machine learning techniques. I also used the Twitter API to gather data for model training and validation. Furthermore, I deployed the solution on the web using Flask, which made the model available to the end-users. As a result of our efforts, we submitted a paper to the SemEval-2020 conference, describing our methodology and achievements. Keywords: Python, NLP, Deep Learning, BERT, CNN, Bidirectional GRU, Machine Learning Web App, Twitter API, Flask

As a participant in this program, I gained a comprehensive understanding of the basics of data science. The program was structured into five modules: Module 1: Prerequisites Module 2: Data Preparation Module 3: Web Scraping Module 4: Machine Learning Module 5: Deep Learning The overarching goal of the program was to prepare a new generation of data scientists who are equipped to tackle and solve tomorrow's challenges.

Developing an intelligent recommendation system that leverages various machine learning techniques to suggest products that match each customer's interests. My work involved using exploratory data analysis to gain insights into customer behavior and preferences, as well as implementing machine learning algorithms such as logistic regression, random forests and gradient boosting to create the recommendation engine. The end result was a highly accurate and effective product recommendation system that significantly improved customer engagement and satisfaction. Keywords: Machine Learning, Exploratory Data Analysis, Logistic Regression, Naive Bayes, Decision Tree, Random Forest, Neural Network, Product Recommendation System