Hamza I.
Junior Data Annotation Specialist
Open to offersNew to PlatformAbout
Hamza I. is an accomplished Data Annotation Specialist based in Islamabad, Pakistan, with expertise in audio, video, text, and image annotation, crucial for AI training data. With a strong foundation in machine learning and AI model training, gained through an MS in Artificial Intelligence from FAST-NUCES, Hamza has effectively contributed to AI model enhancement at companies like Reves.ai and TriNet Studios. At Reves.ai, he excels in annotating B2B audio sales calls, focusing on speaker diarization, sentiment analysis, and intent classification. While at TriNet Studios, Hamza expertly handled video and image labeling, utilizing object tracking, video annotation, and OCR tasks to create accurate AI training datasets. Proficient with tools such as Appen Connect, Label Studio, and CVAT, he ensures high-quality, consistent annotations in compliance with project guidelines. Fluent in English and Urdu, Hamza offers remote full-time availability on PKT hours, providing essential RLHF evaluation and quality assurance in annotation projects.
Experience
Data Annotation Specialist
Reves.ai · 2025 — PresentAnnotated B2B audio sales call recordings for AI/ML model training, labeling aspects such as speaker turns (diarization), intent, and sentiment (positive/negative/neutral). Executed text annotation tasks, including Named Entity Recognition (NER) and intent classification, to extract structured information from conversational data. Conducted sentiment analysis on spoken and written content to aid in training conversational AI models. Adhered strictly to structured annotation schemas and project guidelines, ensuring high levels of inter-annotator agreement (IAA) while identifying edge cases for QA review. Evaluated AI-generated outputs for accuracy and quality, contributing insights to Reinforcement Learning from Human Feedback (RLHF) processes.Data Annotation Specialist
TriNet Studios · 2023 — 2025Executed frame-by-frame video annotation, including drawing bounding boxes and tracking objects across frames for AI/ML training datasets. Performed image classification by categorizing images into specific classes and labeling visual content according to project guidelines. Conducted object detection annotation, marking multiple objects in images with accurate bounding boxes. Managed OCR tasks by transcribing text from various documents into a structured, machine-readable format. Oversaw quality checks on annotated outputs, ensuring accuracy, completeness, and adherence to tight labeling standards before dataset delivery.
Skills & Expertise
Education
- Master of Science in Artificial IntelligenceFAST-NUCES (National University) · — — 2025
- Bachelor of Business and Information TechnologyVirtual University of Pakistan · — — 2022
Interested in this professional?
Sign in as an employer to save this profile or invite Hamza I. to a job.
Sign in as an employer