Hedi El Asmi

Data Analyst & BI Consultant at Avidia Software

Subject : An integrated HR cloud solution that streamlines HR, Payroll, Time, and Talent management. Real-time data analytics within the system to enhance decision-making and operational efficiency.
Tasks :
• Working with Power BI to create and elaborate reports and dynamic dashboards from Azure Cloud.
• Designed and implemented efficient data models in SQL and Power BI for improved data analysis and visualization.
• Created visually appealing and insightful dashboards to present key business metrics and KPIs.
• Applied advanced data transformation techniques to clean, reshape, and prepare data for analysis.
• Implemented complex DAX formulas for creating calculated columns and measures.
• Identified and resolved issues related to data connectivity, model inconsistencies, and visualization errors.
• Successfully delivered Power BI dashboards on the platform for our clients using Angular.
• Providing client support through offering clear explanations for dashboards and reports.
• Developed and implemented customized functionalities, with a focus on Personalized report generation for clients, using .NET & Angular.

Technical Environment :
Microsoft Power BI / Microsoft Excel / Angular & TypeScript / .Net Framework / SQL Server / JavaScript & JQuery / Github / Azure DevOps

Data Scientist at Andex - Paris

Subject : The developed solution allows us to, first of all, scrape the web and download job offers from one or many fields. These offers will then be processed, cleaned and trained according to supervised and unsupervised classification models to result a categorization of the job descriptions in human resources and make predictions of the best possible choices for a given CV.
Tasks :
• Scraping of data according to the profession and location of different fields (Data Science, Finance ...) from the following websites: Indeed, Glassdoor ...
• Deep cleaning of data extracted from the web.
• Data augmentation with "Easy Data Augmentation"
• Use of Natural Language Processing models to correct and translate job descriptions.
• Use of Embedding techniques: Word2Vec, Bag-Of-Words, Tf-IDF, Doc2Vec
• Machine Learning: Supervised learning models: Logistics Regression, SVM, Naïve Bayes ..
• Machine Learning: Unsupervised learning models: K-means ++, Generation of class clusters with keywords.
• Use of a deep learning model: LSTM on textual data.
• Development of a Business report to determine the costs of deploying models on Azure Cloud and Azure Machine Learning.
Results :
• Achievement of F1 Prediction Score with 93% for supervised models and 90% with unsupervised model.
• Prediction of jobs according to the skills or experience of a profile.
Technical Environment : Python 3 + librairies / GitHub / Jupyter Notebook

Data Scientist Intern

Subject : The solution is a website that simulates the use of an algorithm who extracts information from CVs (Word and Pdf) and store them in a database.
Tasks :
• Creation of a script that extracts all the information from a CV in French or English in PDF and Word formats.
• Storage of sorted information in a PostgreSQL database
• Visualize the CV's information from the website
Technical Environment : Python 3 / PostgreSQL / Flask / Html

Mobile Developer Intern

Subject : A mobile cross-platform application of Avidia Software that works on IOS and Android. The mobile application is directly connected with the profile of the user and allows him to stay updated and request leaves.
Tasks :
• Creation of graphic models and a dynamic prototype with JustInMind Prototyper
• Integration of API Rest web services from the web application.
• Development of the various functionalities of the application: Enable employees to engage, communicate and process time off / human resources requests.
Technical Environment : Xamarin / XAML / C# / JustInMind Prototyper / REST