Sakif M.

Key Skills

Software

Other

Top Subject Matter

Gesture recognition using multi-channel EMG (numerical + image labeling)

Crop disease recognition (image data labeling)

Top Data Types

Image

Top Task Types

Classification

Diagnosis

Freelancer Overview

Contrastive and Generative Self-Supervised Learning on Multi-Channels EMG (Robust Gesture Classification). Brings 1+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Bachelor of Science, East West University (2022). AI-training focus includes data types such as Medical, DICOM, and Image and labeling workflows including Classification and Diagnosis.

Labeling Experience

Cucumber Disease Recognition (Machine Learning & Computer Vision)

OtherImageDiagnosis

Developed a machine learning and computer vision disease recognition project by performing strict image data labeling to establish accurate ground truth. Labeled crop images to support defect visualization and disease identification for robust model training. Emphasized maintaining high-quality annotations to improve model reliability and detection outcomes.• Conducted image labeling to create labeled datasets for plant disease and defect recognition.• Established accurate ground truth annotations under strict labeling procedures.• Supported training of models for visually finding defects and recognizing diseases in crops.• Performed dataset preparation steps to ensure labeled images matched disease classes for learning.

2022 - Present

Contrastive and Generative Self-Supervised Learning on Multi-Channels EMG (Robust Gesture Classification)

OtherClassification

Worked on robust gesture classification by creating high-quality annotated datasets from multi-channel EMG inputs for training self-supervised models. Performed end-to-end numerical and image labeling to prepare aligned labels for contrastive and generative learning pipelines. Focused on producing accurate ground truth annotations that improved downstream gesture recognition performance.• Prepared numerical labels and image-based labels for multi-channel EMG signals across channels.• Curated and validated annotated datasets to support contrastive and generative training objectives.• Ensured label consistency between input channels and model targets during data preparation.• Generated usable training artifacts (labeled samples) for self-supervised learning on gesture classes.

2022 - Present

Education

Work History