Ing. Manel BELHADJ

Engineered a custom machine learning model based on GPT-2 to predict sentiment in the Tunisian dialect, significantly improving the platform's ability to handle diverse languages and dialects. Developed and maintained robust web scrapers for real-time social media monitoring, enabling the collection of valuable data on behavior, sentiment, and engagement metrics. Built scalable backend APIs using Django and Express to process scraped data, calculate KPIs, and generate insights on social interactions.

Developed full-stack features for a web application managing companies subscriptions, including creating scalable APIs with NestJS, GraphQL, and Prisma for efficient data management. Designed and built user interfaces using React and shadcn, ensuring a seamless and intuitive user experience. Integrated third-party services like Twilio for SMS notifications and SendGrid for email notifications. Implemented a real-time notification system using sockets and GraphQL WS. Actively participated in code reviews.

AWS re/start program is a training program on IT fundamentals, AWS cloud and professional skills. - Built IT fundamental skills, including: working knowledge of Linux OS, writing scripts in Python and Shell. - Learned how to apply core AWS services in the areas of compute, storage and networking; including EC2, S3, IAM, VPC, Lambda, Cloud Formation, RDS and Route 53. - Learned how to configure, monitor, and support the suite of core AWS services.

Leveraging Machine Learning to predict accomplishment rate of startups: 1. Web Scraping 2. Data preprocessing and EDA 3. Machine Learning Models 4. Model Deployment Exploring the Impact of Covid-19 on Mental Health in Singapore using NLP: 1. Data scraping from online sources (social media, online forums) 2. Sentiment and empath analysis on text data 3. Data visualization and building dashboard 4. Building a predictive model with text data Skills: Data Mining · NLP · Web Scraping · Machine Learning

Built an AI model to recommend cognitive exercises and training programs given: A subject’s baseline capabilities Stated cognitive improvement goals Available cognitive exercises Subjects demographic information. The AI model is able to learn and refine the program through feedback based on adherence to the program and subsequent outcomes. Skills: PyTorch · Recommender Systems · Deep Learning · System Deployment · Python (Programming Language) · Machine Learning