Salami Olaniyi

Key Skills

Software

CVAT

Label Studio

Top Subject Matter

Computer Vision

Nlp Domain Expertise

Conversational AI

Top Data Types

Image

Top Task Types

Bounding Box

Polygon

Segmentation

Classification

Freelancer Overview

AI Data Specialist (Annotation, Data Labeling, Evaluation Scenario Writer & Voice Actor). Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Label Studio, CVAT, and ElevenLabs. Education includes Bachelor of Science, Olabisi Onabanjo University (2022). AI-training focus includes data types such as Image and labeling workflows including Classification.

Labeling Experience

AI Data Specialist (Annotation, Data Labeling, Evaluation Scenario Writer & Voice Actor)

Label StudioImageClassification

As a freelance AI Data Specialist, I annotated and labeled over 15,000 data instances for computer vision and NLP projects using industry-leading platforms. I also designed detailed evaluation scenarios and crafted technical prompts to test and improve AI model performance. Additionally, I produced professional voice-over audio assets for conversational AI datasets. • Labeled Images and Text for classification and evaluation purposes • Used LabelStudio and CVAT for precise data annotation • Delivered test cases and prompts to enhance AI robustness • Produced audio using ElevenLabs and Audacity for AI training

As a freelance AI Data Specialist, I annotated and labeled over 15,000 data instances for computer vision and NLP projects using industry-leading platforms. I also designed detailed evaluation scenarios and crafted technical prompts to test and improve AI model performance. Additionally, I produced professional voice-over audio assets for conversational AI datasets. • Labeled Images and Text for classification and evaluation purposes • Used LabelStudio and CVAT for precise data annotation • Delivered test cases and prompts to enhance AI robustness • Produced audio using ElevenLabs and Audacity for AI training

2023 - Present

AI Training/Data Labeling

ImageBounding Box

I contributed to an image annotation project focused on object detection for AI model training. Scope: The project involved labeling real-world images to create high-quality training data for computer vision models, with applications in areas such as autonomous vehicles, healthcare, sports analytics, and general object recognition. Specific Tasks: - Performed bounding box annotation on thousands of images, drawing tight rectangular boxes around target objects while following detailed annotation guidelines. - Identified and labeled multiple object classes per image, handling various scenarios including occlusion, varying lighting conditions, and crowded scenes. - Conducted quality reviews on my own and peer annotations to ensure consistency. Project Size: Labeled approximately X,XXX images (or X,XXX bounding boxes) over the project duration. Quality Measures: - Achieved and maintained over 95% accuracy through strict adherence to provided guidelines, including tight bounding box rules (minimal padding, proper overlap with object edges). - Participated in inter-annotator agreement checks and feedback loops to resolve ambiguities. - Used model-assisted pre-labeling where applicable, followed by manual correction for precision.

I contributed to an image annotation project focused on object detection for AI model training. Scope: The project involved labeling real-world images to create high-quality training data for computer vision models, with applications in areas such as autonomous vehicles, healthcare, sports analytics, and general object recognition. Specific Tasks: - Performed bounding box annotation on thousands of images, drawing tight rectangular boxes around target objects while following detailed annotation guidelines. - Identified and labeled multiple object classes per image, handling various scenarios including occlusion, varying lighting conditions, and crowded scenes. - Conducted quality reviews on my own and peer annotations to ensure consistency. Project Size: Labeled approximately X,XXX images (or X,XXX bounding boxes) over the project duration. Quality Measures: - Achieved and maintained over 95% accuracy through strict adherence to provided guidelines, including tight bounding box rules (minimal padding, proper overlap with object edges). - Participated in inter-annotator agreement checks and feedback loops to resolve ambiguities. - Used model-assisted pre-labeling where applicable, followed by manual correction for precision.

2025 - 2026

Education

Work History