Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Hi, I’m Janbout Hakim, a Computer Vision and Machine Learning Engineer focused on end-to-end, real-time emotion recognition and robust computer vision pipelines. I enjoy turning ideas into deployable Python-based solutions and integrating ML models into desktop and full-stack systems. I have hands-on experience with CNN design, data preprocessing, and optimizing real-time performance, using Python, TensorFlow, OpenCV, and backend technologies such as Java (Spring Boot) and Node.js. I’m eager to learn, contribute, and grow within applied AI or software development environments, and I’m excited to bring cross-disciplinary collaboration to projects—from computer vision research to product-facing software.…Hi, I’m Janbout Hakim, a Computer Vision and Machine Learning Engineer focused on end-to-end, real-time emotion recognition and robust computer vision pipelines. I enjoy turning ideas into deployable Python-based solutions and integrating ML models into desktop and full-stack systems. I have hands-on experience with CNN design, data preprocessing, and optimizing real-time performance, using Python, TensorFlow, OpenCV, and backend technologies such as Java (Spring Boot) and Node.js. I’m eager to learn, contribute, and grow within applied AI or software development environments, and I’m excited to bring cross-disciplinary collaboration to projects—from computer vision research to product-facing software.

Janbout Hakim

Data Scientist, AI Engineer, Full Stack Developer, +2





Hi, I’m Janbout Hakim, a Computer Vision and Machine Learning Engineer focused on end-to-end, real-time emotion recognition and robust computer vision pipelines. I enjoy turning ideas into deployable Python-based solutions and integrating ML models into desktop and full-stack systems. I have hands-on experience with CNN design, data preprocessing, and optimizing real-time performance, using Python, TensorFlow, OpenCV, and backend technologies such as Java (Spring Boot) and Node.js. I’m eager to learn, contribute, and grow within applied AI or software development environments, and I’m excited to bring cross-disciplinary collaboration to projects—from computer vision research to product-facing software.…Hi, I’m Janbout Hakim, a Computer Vision and Machine Learning Engineer focused on end-to-end, real-time emotion recognition and robust computer vision pipelines. I enjoy turning ideas into deployable Python-based solutions and integrating ML models into desktop and full-stack systems. I have hands-on experience with CNN design, data preprocessing, and optimizing real-time performance, using Python, TensorFlow, OpenCV, and backend technologies such as Java (Spring Boot) and Node.js. I’m eager to learn, contribute, and grow within applied AI or software development environments, and I’m excited to bring cross-disciplinary collaboration to projects—from computer vision research to product-facing software.

Available to hire

Hi, I’m Janbout Hakim, a Computer Vision and Machine Learning Engineer focused on end-to-end, real-time emotion recognition and robust computer vision pipelines. I enjoy turning ideas into deployable Python-based solutions and integrating ML models into desktop and full-stack systems. I have hands-on experience with CNN design, data preprocessing, and optimizing real-time performance, using Python, TensorFlow, OpenCV, and backend technologies such as Java (Spring Boot) and Node.js.

I’m eager to learn, contribute, and grow within applied AI or software development environments, and I’m excited to bring cross-disciplinary collaboration to projects—from computer vision research to product-facing software.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

Arabic

Fluent

English

Advanced

German

Beginner

Work Experience

Emotion Recognition System Developer (Independent Project) at Independent Project

January 1, 2024 - Present

Built a real-time facial emotion recognition system capable of classifying 7 emotional states from webcam, image, or video input. Constructed a custom CNN model (~2.28M parameters) trained on FER+ & RAF-DB, achieving ~79% validation accuracy. Implemented data preprocessing pipelines, including grayscale conversion, 224×224 resizing, face region extraction, and class-balancing augmentation. Integrated the trained model into a desktop application using Tkinter + OpenCV + MediaPipe, supporting real-time live emotion prediction. Designed head pose estimation and facial orientation compensation to improve robustness in real-world lighting and movement conditions.