Qudus Shittu

Key Skills

Software

CVAT

Remotasks

Scale AI

Toloka

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Label Types

Bounding Box

Segmentation

Freelancer Overview

I am an experienced AI evaluator and data annotation specialist with a strong background in computer science and advanced certifications in AI/ML. My expertise spans high-quality data labeling and evaluation across text, image, audio, and video data, supporting the development and alignment of large language models and computer vision systems. I have a proven track record of improving model accuracy and safety through RLHF, prompt engineering, and detailed error analysis for leading platforms like Remotasks and DataAnnotation.tech. Skilled in Python, Pandas, and Scikit-learn, I excel at multi-modal annotation, NLP labeling, and statistical analysis, ensuring production-grade data integrity and consistency for AI training and benchmarking projects.

Labeling Experience

Audio Recording & Transcription

Scale AIAudioAudio RecordingTranscription

NLP Project: ASR Data Labeling & Transcription Objective Enhanced Automatic Speech Recognition (ASR) models by converting 500+ hours of multi-dialect audio into high-precision, machine-readable datasets for NLP training. Core Technical Tasks - Verbatim Labeling: High-accuracy transcription including speaker diarization and precise timestamping. - Acoustic Tagging: Identified non-speech events (noise/overlap) to optimize Neural Network signal processing. - Dataset Prep: Applied orthographic rules to ensure phonetic consistency for model ingestion. Scale & Quality - Project Size: Processed 500+ hours across Medical, Legal, and Casual domains. - Accuracy: Maintained a 98%+ Accuracy Rate, consistently exceeding Word Error Rate (WER) benchmarks. - Compliance: Adhered to strict style guides and PII data privacy protocols.

NLP Project: ASR Data Labeling & Transcription Objective Enhanced Automatic Speech Recognition (ASR) models by converting 500+ hours of multi-dialect audio into high-precision, machine-readable datasets for NLP training. Core Technical Tasks - Verbatim Labeling: High-accuracy transcription including speaker diarization and precise timestamping. - Acoustic Tagging: Identified non-speech events (noise/overlap) to optimize Neural Network signal processing. - Dataset Prep: Applied orthographic rules to ensure phonetic consistency for model ingestion. Scale & Quality - Project Size: Processed 500+ hours across Medical, Legal, and Casual domains. - Accuracy: Maintained a 98%+ Accuracy Rate, consistently exceeding Word Error Rate (WER) benchmarks. - Compliance: Adhered to strict style guides and PII data privacy protocols.

2022 - 2025

NLP: Text Summarization & LLM Fine-Tuning

Scale AITextText Summarization

NLP: Text Summarization & LLM Fine-Tuning Objective: Sourced and synthesized diverse web-based text (news/articles) to provide "gold standard" training data for Abstractive Summarization models. Key Technical Tasks - RLHF Data Prep: Curated high-density summaries to improve model coherence and reduce "hallucinations." - Dataset Diversity: Sourced across multiple domains to ensure model generalization. - Semantic Alignment: Ensured summary outputs matched the core intent of the source text. Quality & Impact - Verification: Performed rigorous cross-checks for factual accuracy and PII redaction. - Performance: Consistently met high-tier quality audits for linguistic fluency and brevity.

NLP: Text Summarization & LLM Fine-Tuning Objective: Sourced and synthesized diverse web-based text (news/articles) to provide "gold standard" training data for Abstractive Summarization models. Key Technical Tasks - RLHF Data Prep: Curated high-density summaries to improve model coherence and reduce "hallucinations." - Dataset Diversity: Sourced across multiple domains to ensure model generalization. - Semantic Alignment: Ensured summary outputs matched the core intent of the source text. Quality & Impact - Verification: Performed rigorous cross-checks for factual accuracy and PII redaction. - Performance: Consistently met high-tier quality audits for linguistic fluency and brevity.

2023 - 2024

Image Annotation and Segmentation

Scale AIImageBounding BoxSegmentation

Drawing bounding boxes round objects in images and also outlines.

Drawing bounding boxes round objects in images and also outlines.

2021 - 2024

Education

Work History