Adekanye Abdulkabir

Key Skills

Software

Roboflow

Other

Internal/Proprietary Tooling

Top Subject Matter

General Computer Vision

Multiple domains and LLM Feedback

Flutter Software developer

Top Data Types

Image

Text

Audio

Top Task Types

Object Detection

RLHF

Transcription

Bounding Box

Entity Ner Classification

Classification

Freelancer Overview

AI Data Annotator & Trainer (Freelance). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Roboflow, N, and A. A Senior Flutter developer with 6+ years of experience in developing and shipping production-grade applications. AI-training focus includes data types such as Image, Text, and Audio, and labeling workflows including Object Detection, RLHF, and Transcription.

Labeling Experience

Audio Annotator

AudioTranscription

Performed audio transcription and speaker diarisation labelling for multilingual voice-assistant datasets. I ensured strict verbatim accuracy and high-quality labels suitable for automated speech systems. The work required high attention to linguistic detail in multiple languages. • Labelled multilingual speech segments and speakers • Maintained rigorous transcription standards • Adapted quickly to new languages and accents • Adhered to strict QA protocols for audio data annotation

Performed audio transcription and speaker diarisation labelling for multilingual voice-assistant datasets. I ensured strict verbatim accuracy and high-quality labels suitable for automated speech systems. The work required high attention to linguistic detail in multiple languages. • Labelled multilingual speech segments and speakers • Maintained rigorous transcription standards • Adapted quickly to new languages and accents • Adhered to strict QA protocols for audio data annotation

2023 - Present

LLM Response Ranker / RLHF Evaluator

TextRLHF

As part of RLHF pipelines for LLM fine-tuning projects, I evaluated and ranked AI-generated text responses on factual accuracy, helpfulness, tone, and instruction-following. This work informed large datasets for language model refinement. I contributed structured ranking rationales and supported QA. • Ranked and evaluated hundreds of AI-generated responses • Focused on factuality, harmlessness, and instruction adherence • Composed detailed rationales for each ranking • Ensured high standards in dataset consistency and documentation

As part of RLHF pipelines for LLM fine-tuning projects, I evaluated and ranked AI-generated text responses on factual accuracy, helpfulness, tone, and instruction-following. This work informed large datasets for language model refinement. I contributed structured ranking rationales and supported QA. • Ranked and evaluated hundreds of AI-generated responses • Focused on factuality, harmlessness, and instruction adherence • Composed detailed rationales for each ranking • Ensured high standards in dataset consistency and documentation

2023 - Present

AI Data Annotator & Trainer (Freelance)

RoboflowImageObject Detection

As an AI Data Annotator & Trainer, I annotated over 10,000 images across object detection, segmentation, and classification tasks using Roboflow, CVAT, and Label Studio. I maintained strict accuracy standards with over 95% accuracy on all batches, contributing to high-quality datasets. I also created guidelines and QA processes adopted by multiple annotation teams. • Performed image annotation including bounding boxes and segmentation masks • Used Roboflow, CVAT, Label Studio for annotation tasks • Authored detailed annotation guidelines and decision trees • Conducted structured QA reviews for inter-annotator agreement

As an AI Data Annotator & Trainer, I annotated over 10,000 images across object detection, segmentation, and classification tasks using Roboflow, CVAT, and Label Studio. I maintained strict accuracy standards with over 95% accuracy on all batches, contributing to high-quality datasets. I also created guidelines and QA processes adopted by multiple annotation teams. • Performed image annotation including bounding boxes and segmentation masks • Used Roboflow, CVAT, Label Studio for annotation tasks • Authored detailed annotation guidelines and decision trees • Conducted structured QA reviews for inter-annotator agreement

2023 - Present

NLP Annotation – Kaviel Threat Classifier

TextEntity Ner Classification

Curated and annotated a threat intelligence classification dataset for the Kaviel Threat Classifier project in 2024. I defined labeling taxonomy, annotated text for NER and text classification, and finalized guidelines. The dataset was used to fine-tune a RoBERTa-base transformer for 6 category intent recognition. • Labeled cyber threat intelligence texts for NER and classification • Built and reviewed custom dataset for ML training • Defined annotation and label taxonomy guidelines • Evaluated model with ROC AUC and class-wise accuracy metrics

Curated and annotated a threat intelligence classification dataset for the Kaviel Threat Classifier project in 2024. I defined labeling taxonomy, annotated text for NER and text classification, and finalized guidelines. The dataset was used to fine-tune a RoBERTa-base transformer for 6 category intent recognition. • Labeled cyber threat intelligence texts for NER and classification • Built and reviewed custom dataset for ML training • Defined annotation and label taxonomy guidelines • Evaluated model with ROC AUC and class-wise accuracy metrics

2024 - 2024

Image Annotation – Crop Vision Dataset

RoboflowImageBounding Box

Collected and annotated agricultural images using bounding boxes and polygon masks for Crop Vision Dataset. I maintained class balance, validated label consistency, and prepared the data for model training. The annotation workflow was optimised for high quality and mAP performance. • Annotated 3,000+ agricultural images with precise bounding and segmentation • Used Roboflow for labeling and validation • Maintained consistent labeling standards across classes • Delivered training-ready computer vision data for agricultural analysis

Collected and annotated agricultural images using bounding boxes and polygon masks for Crop Vision Dataset. I maintained class balance, validated label consistency, and prepared the data for model training. The annotation workflow was optimised for high quality and mAP performance. • Annotated 3,000+ agricultural images with precise bounding and segmentation • Used Roboflow for labeling and validation • Maintained consistent labeling standards across classes • Delivered training-ready computer vision data for agricultural analysis

2022 - 2022

Show more

Education

Work History